In Kevin Khoa Nguyen v. Barnes & Noble Inc., 2014 U.S. App. LEXIS 15868 (9th Cir. 2014), decided on August 18, 2014, the Ninth Circuit rejected an attempt to bind a consumer to an arbitration clause found in an online terms of use agreement not affirmatively “click accepted” by the consumer but readily accessible through a hyperlink at the bottom left of each page on the subject website.

The case arose from a “fire sale” by defendant Barnes & Noble of certain discontinued Hewlett Packard TouchPads. Plaintiff Nguyen had ordered two of the TouchPads, but received a notice from Barnes & Noble the following day that his order had been cancelled due to unexpectedly high demand. Nguyen sued Barnes & Noble in California Superior Court on behalf of himself and a putative class, arguing that he was forced to buy a more expensive tablet instead.

Barnes & Noble, after removing the suit to federal court, moved to compel arbitration under the Federal Arbitration Act, arguing that, by using the Barnes & Noble website, Nguyen had agreed to an arbitration clause contained in Barnes & Noble’s Terms of Use. Nguyen responded that he could not be bound to the arbitration clause because he had no notice of and did not consent to the Terms of Use. Barnes & Noble countered that the placement of the Terms of Use hyperlink on its website had given Nguyen constructive notice of the arbitration clause.

Continue Reading To Click or Not to Click? Ninth Circuit Rejects Browsewrap Arbitration Clause

Website operators often take for granted the enforceability of their websites’ terms of service. In a recent order issued in a case from the Central District of California, Nguyen v. Barnes & Noble, Inc., Judge Josephine Tucker reminds us that such presumptions are not necessarily correct: terms of service that do not require an affirmative manifestation of assent from a website user may not always be upheld in court.

Many website operators, particularly Internet retailers and operators of ecommerce sites, use “clickwrap” (or “clickthrough”) agreements to govern use of their sites. With clickwrap agreements, the website operator typically presents its standard terms of use and then requires the user to click an “Accept” or “I Agree” button. By clicking the button, users affirmatively manifest their intent to be bound by the terms. Other website operators use “browsewrap” agreements—terms of agreement that are usually accessible through a hyperlink at the bottom of a web page. Although, as a practical matter, few people actually read them, browsewraps are also widely used.

Both clickwraps and browsewraps are contracts of adhesion in legal parlance. That is, they are contracts that are offered on a “take it or leave it” basis with no opportunity for negotiation. A user who does not wish to be bound by the proffered terms can click “Do Not Accept” or, for a browsewrap, simply leave the website. On the other hand, a user who is willing to be bound can indicate such assent by clicking “I Accept” or by continuing to browse the website. Reasonable people may disagree regarding whether these actions truly manifest a user’s assent to be bound by the relevant contract terms, but courts have frequently upheld the enforceability of both clickwrap and browsewrap terms of use (subject, of course, to the unconscionability concerns raised by any contract of adhesion). As discussed in the remainder of this article, however, browsewrap terms of use often encounter a greater degree of scrutiny from courts due to the lack of any affirmative acceptance by users.

The enforceability of browsewrap terms of use has been held to depend on whether a website user has knowledge—either actual or constructive—of the applicable terms, because users cannot agree to be bound by terms unless they know what those terms are. Courts considering browsewrap enforceability issues often grapple with the question of whether the defendant was given notice of the applicable terms sufficient to impute such knowledge. For example, in, Inc. v. Verio, Inc., the court determined that numerous and repeated queries by an automated software program were sufficient to show that Verio knew of, and was bound by,’s terms (although Verio had also admitted that it had actual knowledge of the terms). On the other hand, in Ticketmaster Corp. v., Inc., on the other hand, the court held that a small link to terms of use that was visible only if the user scrolled down to the bottom of the web page was insufficient to establish notice. But, three years later, the same court (in the same case, no less) ruled that more prominent notice on the site’s home page was adequate notice. While a court’s determination of sufficient notice may vary in each case, it is clear that the more readily available and conspicuous browsewrap terms of use are, the more likely it is that a court will find that the user knew of, and was bound by, the terms.

That brings us to Nguyen v. Barnes & Noble, Inc. In Nguyen, the plaintiff’s claims arose from a Barnes & Noble promotion that offered computer tablets at a discounted price. Although Nguyen submitted an order to purchase a tablet at the promotional price, Barnes & Noble canceled his order the next day, citing an oversale of its tablet inventory. As a result, Nguyen alleged that he was “forced to rely on substitute tablet technology, which he subsequently purchased . . . [at] considerable expense.” In April 2012, Nguyen filed suit, alleging various consumer protection violations, including false advertising, unfair competition, and breach of contract, under California and New York law. Barnes & Noble then moved to compel arbitration based on an arbitration clause included in its website’s browsewrap terms of use. The question before the court was whether, given the existing facts, the arbitration clause was enforceable against Nguyen.

The court ultimately held that the arbitration clause was not enforceable because the terms of use agreement itself was not enforceable. According to Judge Tucker, Barnes & Noble’s website terms of use could not bind Nguyen because Barnes & Noble “did not position any notice even of the existence of its ‘Terms of Use’ in a location where website users would necessarily see it, and certainly did not give notice that those Terms of Use applied, except within the Terms of Use” (emphasis in original). Due to this lack of adequate notice, Nguyen did not know and, in Tucker’s view, should not necessarily have known of Barnes & Noble’s terms of use. Because Nguyen did not have knowledge of the terms, he could not be bound by them. Therefore, Barnes & Noble could not compel arbitration in its dispute with Nguyen.

In light of Nguyen and the other cases discussed above, website operators should consider using clickwraps that require affirmative acceptance where possible, rather than relying on browsewraps to enforce their terms of use. A simple click can be the difference between an agreement’s being found enforceable or not. For ecommerce sites or any site that requires registration prior to use, clickwraps are relatively easy to implement—for example, at the point of purchase or when the user registers—without negatively affecting the user experience. Best practices for clickwraps include presenting terms of service before payment, allowing for easy reading of all terms, allowing users to print or save a copy of the terms, offering a prominent option to decline the terms, providing an easy way for users to find the terms on the site at any time after payment or registration, and giving users notice of (and requiring users to accept) any updates and changes to the terms of use.

For other sites, including some social media sites, the story may differ. Many social media sites—for example, Pinterest, Twitter, and YouTube—allow users to access at least some content and functionality without registering. With sites such as these, there may be no real opportunity to obtain affirmative acceptance of terms of use without degrading the user experience, so a clickwrap is simply not a practical option. For operators of such websites, the most important lesson of Nguyen and the other cases discussed above is that the question of enforceability often turns on whether the user has sufficient notice of the terms of use. Thus, website operators can increase the likelihood that their terms of use will be enforced if links to such terms are prominently displayed, preferably “above the fold” so that a user will be able to see the link without scrolling down the page. As Nguyen and the other cases illustrate, an operator who places links to terms of use in a tiny font buried at the bottom of a page may be in for an unpleasant surprise if those terms ever need to be enforced.

In two recent decisions issued within a day of each other, two influential federal courts limited the scope of three important federal laws used to prosecute criminal conduct involving computers.  On April 10, 2012, the Ninth Circuit limited the scope of criminal liability for prosecutions under the Computer Fraud and Abuse Act, and on the following day the Second Circuit sharply limited the scope of the National Stolen Property Act and the Economic Espionage Act of 1996.  Together, these decisions indicate a reluctance to accept prosecutors’ expansive views of the reach of federal criminal laws with respect to computer usage, and the Ninth Circuit’s decision in particular may have far-reaching implications for the enforceability of website terms of service and employee policies in the civil context.

The Ninth Circuit’s decision was issued en banc in United States v. Nosal upholding the district court’s dismissal of David Nosal’s indictment for violations of the Computer Fraud and Abuse Act (“CFAA”).  Nosal had worked for an executive search firm and left to start a competing business.  He convinced several of his former colleagues to help him by accessing and then transferring to him source lists, names, and contact information from the firm’s confidential database.  The former colleagues were authorized to access the database, but the firm had a policy forbidding the disclosure of confidential information.  The government charged Nosal with violating 18 U.S.C. § 1030(a)(4) by aiding and abetting the former colleagues in “exceed[ing] authorized access” to the firm’s computers with intent to defraud the firm.

Nosal moved to dismiss the CFAA counts, arguing that the statute was meant to target hackers and not those who accessed a computer lawfully but then misused information obtained from such access.  The district court agreed, and the government appealed.  In a panel decision issued in 2011, the Ninth Circuit reversed the district court, holding that an employee “‘exceeds access’ under § 1030 when he or she violates the employer’s computer access restrictions — including use restrictions.”  The en banc court found otherwise, holding that “‘exceeds authorized access’ in the CFAA is limited to violations of restrictions on access to information, and not restrictions on its use.” (Emphasis in original.)  To hold otherwise, the court reasoned, would make federal crimes out of “minor dalliances” like playing games or shopping online, if such activities were prohibited by an employer’s computer-use policy.  The court observed:  “Employer-employee and company-consumer relationships are traditionally governed by tort and contract law,” and to interpret the CFAA to apply to use restrictions “allows private parties to manipulate their computer-use and personnel policies so as to turn these relationships into ones policed by the criminal law.”  This would implicate “[s]ignificant notice problems.”  Although the government argued that it would not prosecute minor violations of the law, the court found that “we shouldn’t have to live at the mercy of our local prosecutor.”

The Second Circuit’s decision in United States v. Aleynikov, issued on April 11, 2012, limits the reach of computer crime prosecutions under the National Stolen Property Act (“NSPA”) and the Economic Espionage Act of 1996 (“EEA”).  Sergei Aleynikov was convicted of violating both acts based on his theft and transfer of his company’s proprietary source code.  Aleynikov was a computer programmer at Goldman Sachs, where he developed source code for the company’s proprietary high-frequency trading (“HFT”) system.  Goldman’s policies bound Aleynikov to keep the firm’s proprietary information confidential and barred him from taking or using it when his employment ended.  Aleynikov accepted an offer from a new company that was looking to develop its own HFT system.  On his last day at Goldman, Aleynikov uploaded source code for Goldman’s HFT system to a server in Germany, which he then downloaded to his home computer for use at his new job. 

Aleynikov was sentenced to 97 months in prison.  He appealed, arguing that the district court should have dismissed his indictment for failure to state an offense.  The Second Circuit reversed his conviction on both counts, finding that his conduct did not constitute an offense under either statute.  (Aleynikov has also been charged with a criminal violation of the CFAA, but the district court had dismissed that charge on the ground that “authorized use of a computer in a manner that misappropriates information is not an offense” under the act.  This ruling predates the similar en banc Nosal decision discussed above, and the government did not appeal the ruling.)

The NSPA criminalizes transmittal of a stolen “good” in interstate or foreign commerce.  The Second Circuit held that source code is not a “good,” and therefore, “the theft and subsequent interstate transmission of purely intangible property is beyond the scope of the NSPA.”  The court “decline[d] to stretch or update statutory words of plain and ordinary meaning in order to better accommodate the digital age.”  Significantly, the court noted that a different conclusion might apply if the stolen source code had been removed from Goldman’s premises on a tangible item, like a CD or flash drive, instead of having been stolen through uploading to an off-premises server.

The EEA prohibits the unauthorized downloading, uploading, transmitting, or conveying of trade secrets related to or included in a product that is produced for or placed in interstate or foreign commerce, with the intent to convert the trade secret, while intending or knowing that the offense will injure the owner of the trade secret.  On this count, the Second Circuit held that Goldman’s HFT system was neither “produced for” nor “placed in” interstate commerce because Goldman had no intention of selling or licensing the system and, in fact, “went to great lengths to maintain the secrecy of its system.” 

Although neither the NSPA nor EEA provides for a private right of action, we think it is possible the rationales of these decisions could influence civil litigation involving misuse of an employer’s computer system, including, in particular, civil litigation under the CFAA based on violations of website terms of service or employee policies.  For examples of previous such cases, see, e.g., Am. Online, Inc. v. LCGM, Inc. and EF Cultural Travel BV v. Explorica, Inc. In most of these cases, it appears that the defendant was authorized to access the website or system in question, but misappropriated the data on those websites or systems.  In addition to limiting criminal exposure, the Ninth Circuit’s interpretation of “exceeds authorized access” in Nosal may be construed to undermine this basis for a civil suit.  Watch these pages for further reports on these issues.

In the recent online contracting case of Fteja v. Facebook, Inc., a New York federal court held that a forum selection clause contained in Facebook’s Statement of Rights and Responsibilities (the “Terms”) was enforceable because the plaintiff assented to the Terms when registering to use Facebook.  The court’s analysis and holding followed the recent trend of de-emphasizing the distinction between “clickwrap” and “browsewrap” agreements and instead focusing on whether the user was provided  with actual or constructive notice of the agreement’s terms and conditions.  In this case, the result turned on whether Facebook’s Terms were reasonably communicated to the plaintiff prior to his use of the site.

The plaintiff, an active Facebook user, brought the action against Facebook in New York state court asserting that Facebook disabled his account without justification and for discriminatory reasons.  He claimed that the disabling of his account hurt his feelings, inflicted emotional distress and assaulted his good reputation among his friends and family.

Facebook removed the lawsuit to New York federal court on the basis of diversity of citizenship, and then moved to transfer the action to federal court in Northern California, citing the forum selection clause in the Terms.  Facebook argued that because the plaintiff clicked through Facebook’s registration page and expressly acknowledged that he read and agreed to the Terms (including the forum selection clause), the Terms were valid and enforceable.  The plaintiff responded that there was no proof that he agreed to the forum selection clause and that he did not remember agreeing to the Terms.

The court reviewed Facebook’s registration process, noting that after a new user provides his or her personal information and clicks an initial “Sign Up” button, he or she is directed to a security page that requires the new user to input a series of letters and numbers.  Below the box where the new user enters the letter/number combination, the page displays a second “Sign Up” button that is immediately followed by the phrase: “By clicking Sign Up, you are indicating that you have read and agree to the Terms of Service.”  The phrase “Terms of Service” is underlined, indicating that it is hyperlinked to the Terms.

After this review of Facebook’s registration process, the court then described the historical development of online contracting law, referencing the, Inc. v. Verio, Inc., Specht v. Netscape Communications Corp., and Hines v., Inc. decisions, and the importance of establishing mutual manifestation of assent.  Following this discussion, the court pointed out that Facebook’s Terms are “somewhat like a browsewrap agreement in that the terms are only visible via a hyperlink, but also somewhat like a clickwrap agreement in that the user must do something else – click ‘Sign Up’ – to assent to the hyperlinked terms,” and that, unlike some clickwrap agreements, a new Facebook user can click to assent whether or not he or she has been presented with the Terms.  Finally, the court looked at the plaintiff’s level of sophistication and stated that an Internet user whose social networking was so prolific that losing Facebook would cause him mental anguish should understand that the hyperlinked phrase “Terms of Service” really means “Click Here for Terms of Service,” thereby establishing constructive knowledge of the Terms.

The court concluded that the plaintiff’s registration for Facebook by clicking the “I accept” button constituted his assent to the Terms (including the forum selection clause) even though he may not have actually reviewed the hyperlinked Terms.  The court then, after considering the public policy ramifications of the transfer decision, held that the forum selection clause was enforceable and directed the action to be transferred to federal court in Northern California.

Key Take Aways.  While the Fteja v. Facebook, Inc. case illustrates that U.S. courts may enforce a hybrid browsewrap/clickwrap agreement even where the user does not have actual knowledge of the terms and conditions, the safest approach for a website operator is to structure its online terms of service as a traditional clickwrap agreement that requires users to scroll through the terms and conditions and then click an “I accept” button.  In situations where this structure is not commercially reasonable, the following tips can be used to help establish user assent under U.S. law through constructive knowledge of the terms and conditions of an online agreement:

  • Prominent Notice:  Include a prominent notice that cannot be skipped by users; such notice ideally should state that the use of service is subject to the hyperlinked terms of service.  Such notice should be provided in reasonably large font and contrasting colors that do not blend into the website’s background.  If possible, include an “I accept” button next to the notice.
  • Easy Access/Full Disclosure:  Provide easy access to the full text of the terms of service via a clearly identifiable hyperlink that links to a downloadable and printable version of the terms of service.  The hyperlink should be provided next to the notice and an “I accept” button (if any).
  • Readability:  Structure and phrase the terms of service so that they can be reasonably understood by users based on their anticipated level of sophistication.
  • Highlight Important Terms:  Make sure that any particularly important terms are clearly identifiable and not hidden.  If the website operator is especially concerned about an issue (e.g., enforceability of the limitation of liability provision of the terms of service), consider expressly referencing the concern as part of the general notice (e.g., “By clicking Sign Up, you are indicating that you have read and agree to the Terms of Service, including the limitations on vendor’s liability described therein”).

Twitpic, a user-generated content service that simplifies the process of sharing photographs and other media through Twitter, came under fire earlier this year for changes to its Terms of Service that appeared to dramatically expand the rights granted to Twitpic by its users— and that were described by some media outlets as a “copyright grab.”

Twitpic’s original Terms of Service stated that “all images uploaded are copyright their respective owners,” and that by uploading photos to Twitpic, users gave Twitpic “permission to use or distribute” such photos on and affiliated sites.  As highlighted in Eric Goldman’s Technology & Marketing Law Blog, Twitpic revised its Terms of Service without notice on May 4, 2011, to state that users who share content via Twitpic “may not grant permission to photographic agencies, photographic libraries, media organizations, news organizations, entertainment organizations, media libraries, or media agencies to retrieve from Twitpic for distribution, license, or any other use, content [users] have uploaded to Twitpic.” Goldman’s blog notes that this language suggested that, if a user posted content through Twitpic, the user could not then license that content to other third parties, even though the license grant to Twitpic was putatively nonexclusive.  That said, the prohibition only applied to retrieval of the uploaded content “from Twitpic” for further use.  Whatever its intent, the offending language was deleted in Twitpic’s next revision to its Terms of Service, which came six days later along with a clarification from Twitpic regarding its revised terms.

Despite the various changes to Facebook’s Guidelines, one fundamental principle remains the same:  The Guidelines continue to govern both communication about, and administration of, promotions on Facebook.

The further-revised (and currently governing) version of Twitpic’s Terms of Service states that “all content uploaded to Twitpic is copyright the respective owners,” and clearly specifies the licenses that Twitpic users are granting.  The first licenses are to Twitpic, including both a broad license similar to the one found in Twitpic’s original Terms of Service (“permission to use or distribute [the user’s] content on or affiliated sites”), along with a more detailed nonexclusive, sublicensable license that permits Twitpic to use, reproduce, distribute, prepare derivative works of, display and perform a user’s uploaded content in connection with the Twitpic service and Twitpic’s, its affiliates’ and successors’ businesses, including for “promoting and redistributing” Twitpic’s service and derivative works of the service, in any format and through any channel.

The second license is a grant to other users of Twitpic, giving them an express right to use, distribute, display and perform a user’s uploaded content “through the functionality of [Twitpic] and under these Terms of Service.” (These licenses terminate within a commercially reasonable time after a user removes or deletes his or her content from Twitpic, however, any sublicense granted prior to such termination “may be perpetual and irrevocable.”)

The licenses to Twitpic itself are very similar to the ones demanded by most services built around user-generated content—and the inclusion of a license-back of each user’s content to the general user population, albeit restricted to Twitpic’s own functionality, addresses an issue that is not often raised in website terms and conditions.  Still, other language in Twitpic’s Terms of Service, when combined with these broad grants, has stirred concerns among Twitpic users.  Although “it is not acceptable to copy or save another user’s content from Twitpic and upload it to other sites for redistribution and dissemination,” the terms also state that in order “to publish another Twitpic user’s content for any commercial purposes [other than simple ‘Retweets’] … whether online, in print publication, television, or any other format, you are required to obtain permission from Twitpic in advance of said usage and attribute credit to Twitpic as the source where you have obtained the content” (emphasis added).  So it appears that the Terms of Service both prohibit users’ commercial use of other user’s content, and expressly contemplate such use so long as that use is controlled by Twitpic (rather than the uploading user) in terms of both approval and functionality.

More recently, it was reported that World Entertainment News Network (“WENN”) and Twitpic had reached an agreement making WENN Twitpic’s exclusive photo agency, which, according to executives, was intended to facilitate legitimate, authorized use of Twitpic images as part of breaking news and entertainment stories.  Some users are reportedly worried that they will not be able to opt out of the WENN/Twitpic arrangement, even though WENN’s CEO has stated that only a very small number of celebrities’ images will be distributed via WENN—and according to The New York Times, certain celebrities whose pictures were to be distributed by WENN have stopped using Twitpic.  Still, despite much sound and fury from the blogosphere, no mass exodus from the popular content-sharing service has materialized as a result of these developments.

In our February 2011 issue of Socially Aware, we reported that, at the end of 2010, Facebook had revamped its Promotions Guidelines (the “Guidelines”) to eliminate the requirement that approval be obtained from Facebook prior to offering sweepstakes, contests or similar promotions in connection with one’s Facebook page.  More recently, Facebook substantially streamlined its Guidelines, which now focus less on specific points of legal compliance and more on how Facebook features and functionality may and may not be used in connection with promotions.  Here are some highlights from the revised Guidelines:

Legal Compliance – Attorneys who advise clients on contests and sweepstakes are well aware of the difficulties in navigating the labyrinth of state, federal and foreign laws that govern such promotions.  Earlier versions of the Guidelines took a proactive approach to ensuring that sweepstakes, contests and similar promotions appearing on Facebook pages complied with applicable laws, by including long lists of prohibitions and restrictions against matters such as making sweepstakes available to individuals in specific jurisdictions (e.g., Belgium, Norway, Sweden or India), making promotions available to those under 18 years of age, and offering types of prizes (such as alcohol, tobacco, dairy, firearms and prescription drugs) that could raise general or promotion-specific legal issues.  The revised Guidelines have eliminated most of these specific restrictions, and now provide that promotion operators themselves are “responsible for the lawful operation of [their promotions], including the official rules, offer terms and eligibility requirements (e.g., age and residency restrictions), and compliance with regulations governing the promotion and all prizes offered in connection with the promotion (e.g., registration and obtaining necessary regulatory approvals).” The Guidelines also clarify that compliance with the terms and conditions of the Guidelines does not, in itself, make a promotion lawful.  Facebook’s new approach is a valuable reminder for promotion operators to seek legal guidance when designing, drafting official rules for, and administering sweepstakes, contests and promotion, whether on social media platforms or otherwise.

Use of Facebook Features and Functionality – In contrast to their new approach to legal compliance, the revised Guidelines now provide more detail on how promotion operators may and may not use Facebook features and functionality in connection with their promotions.  (Some of this detail had been included in “examples” in the prior version of the Guidelines, but may not have been clear from the body of the Guidelines.) A few key examples:

Voting – For years, Internet-based contests have relied on public judging/voting on participants’ entries to determine finalists and winners.  Section 5 of Facebook’s revised Guidelines provide that promotion operators “must not use Facebook features or functionality, such as the Like button, as a voting mechanism for a promotion.” This shifts the burden of supplying a usable and accurate voting mechanism onto promotion operators themselves.  We note that Section 1 of the Guidelines requires that all promotions on Facebook “must be administered with Apps on”, so it is unclear whether a voting mechanism within a custom app may be used notwithstanding this blanket prohibition.

•  Means of entry vs. conditions for entry – The revised Guidelines make a subtle but important distinction between, on one hand, using Facebook features as a mechanism for registering for or entering a promotion, and on the other hand, conditioning registration or entry on taking a particular Facebook action.  Per Section 3 of the revised Guidelines, promotion operators are flatly prohibited from using Facebook features or functionalities (e.g., “liking” a Page, posting an item of content or joining a group) as the actual means of entry for a promotion, without exception.  On the other hand, per Section 4 of the revised Guidelines, promotion operators are prohibited from “condition[ing]” registration for or entry in a promotion on taking actions using Facebook features or functions, but with an important exception—registration or entry may be conditioned on a user “liking a Page, checking in to a Place, or connecting to [the promotion operator’s] app.” So, for example, liking a Page can serve as a requirement for participation in a promotion, but it cannot be the way for a participant to enter the promotion.

•   Notifying winners – A promotion’s official rules typically state how and when winners will be notified.  Section 6 of the revised Guidelines now prohibits promotion operators from notifying winners “through Facebook, such as through Facebook messages, chat, or posts on profiles or Pages.”  (Again, it is unclear whether winners may be notified through a promotion’s app on the Facebook platform.)

Despite the various changes to Facebook’s Guidelines, one fundamental principle remains the same:  The Guidelines continue to govern both communication about, and administration of, promotions on Facebook.  And these two terms are sweeping in scope.  “Communication” includes “promoting, advertising or referencing a promotion in any way on Facebook, e.g., in ads, on a Page, or in a Wall post,” and “administration” includes “the operation of any element of the promotion, such as collecting entries, conducting a drawing, judging entries, or notifying winners.” This means that even mentioning your promotion on Facebook is subject to the Guidelines.  So, before launching your next promotion involving Facebook—whether you are actively operating the promotion on Facebook, or merely referencing an independently run promotion—remember to review the current Guidelines to ensure compliance.

The Superior Court of New Jersey recently revisited the enforceability of online contracts and the importance of how terms and conditions are displayed on websites, in Hoffman v. Supplements Togo Management LLC, et al. In so doing, the court addressed a line of cases reaching back to the Second Circuit’s 2002 landmark decision in Specht v. Netscape, where Circuit Judge (now Justice) Sotomayor wrote that, unless a reasonably prudent Internet user would have learned of and unambiguously assented to terms governing an online commercial transaction, an online contract cannot be formed.  In Hoffman, the court seized the opportunity to clarify how the law’s view of a “reasonably prudent Internet user” has evolved over the intervening nine years in light of the rapid growth in Internet use and online transactions.  As it turns out, the answer is . . . not by much.

The plaintiff in Hoffman, an attorney with an alleged history of suing online retailers for deceptive practices, purchased a dietary supplement called “Erection MD” through a website operated by defendant Supplements Togo Management LLC (“Togo”).  The product in question was advertised on Togo’s site with a variety of claims, such as, “Enhances Sex Drive,” “Maximum Performance,” “Instantly Boost Testosterone Levels,” and “Ultimate Stamina.” Four days after receiving his shipment, the plaintiff filed a lawsuit in the Superior Court of New Jersey alleging violations of New Jersey’s Consumer Fraud Act (“CFA”) and claiming that Togo made false and exaggerated representations about the product that allegedly lacked scientific and objective support.  (New Jersey’s CFA essentially requires advertisers to substantiate with written proof any claims made concerning “the safety, performance, availability, efficiency, quality or price of the advertised merchandise,” and to keep such written proof on file for at least 90 days after the effective date of the advertisement.) In lieu of filing an answer to the complaint, Togo moved to dismiss Hoffman’s suit, arguing that Hoffman failed to state a claim under the CFA and was barred from suing Togo in New Jersey in light of the forum selection clause contained in Togo’s “website disclaimer,” which only permitted actions to be brought in Nevada.  The lower court, in addressing whether the clause was enforceable, dismissed Hoffman’s suit on the grounds of improper forum.

On appeal, the Hoffman court focused on the same key principles of notice and assent discussed in Specht, and in particular, on whether “a reasonably prudent [person] in these circumstances would have known of the existence of [the] license terms.” In this case, the disclaimer containing the forum selection clause was displayed “below the fold” (that is, on a “submerged” portion of the website to which a visitor needed to scroll down in order to see).  Hoffman stated that when he visited Togo’s website, Erection MD was the first product displayed, listed among other Togo products and supplements and appearing next to a box that read “ADD TO SHOPPING CART,” and that when he clicked to add the product to his cart, he was taken directly to the site’s checkout page.  Hoffman argued that because subsequent pages—including the one on which he consummated his purchase—did not contain the same disclaimer, he was never put on notice of those additional terms and, therefore, that Togo’s forum selection clause was unenforceable.  Applying New Jersey precedent, the court agreed with Hoffman and overturned the lower court’s dismissal.  The forum selection clause was ruled “presumptively unenforceable,” on the grounds that it was “proffered unfairly, or with a design to conceal or de-emphasize its provisions.” Persuaded by Hoffman’s argument, the judge emphasized that because the forum selection clause was “submerged” on the web page that listed Togo’s products, it was “unreasonably masked from the view of the prospective purchasers because of its circuitous mode of presentation,” which prevented Hoffman (or any customer) from being put on notice.

The analysis in Hoffman mirrors Specht where Justice Sotomayor noted that the fact that an unexplored portion of a web page could contain additional terms and conditions, does not mean that a reasonably prudent Internet user should assume the existence of—or be compelled to look for—those terms.  Rather, it should be the website operator’s responsibility to put Internet users on notice of applicable terms and to obtain their assent, in order to preserve the integrity and credibility of electronic “bargaining” and mutual assent necessary to establish a contract.  In applying the same logic, the court in Hoffman signaled that the view of what an Internet user (whether or not he or she is an attorney) should be responsible for today has not changed much since Specht, despite the fact that the majority of Internet  users have made purchases online.

Similar to previous clickwrap cases (including Specht), Judge Sabatino also made a point of noting that “if defendants establish on remand that Hoffman had actually read the forum selection clause before purchasing the product,” his ruling on the enforceability of the clause might have been different.  Additionally, on the issue of assent, the Hoffman court stopped short of ruling (as Hoffman had argued) that a website user must be made to expressly click an “I Agree” button or check-box in order to form a binding agreement; although the user’s “unambiguous manifestation of assent” is required, New Jersey’s courts, like others, remain hesitant to prescribe the means or technology that a website operator needs to use to obtain that assent.  (The Hoffman opinion did not address the fact that Togo’s website disclaimer was a browsewrap rather than a clickwrap agreement.) Hoffman is a reminder to website operators everywhere of the continuing importance of highlighting website terms and conditions, and making sure that visitors are on notice that their activities— including purchases—are governed by those terms.  As the line of clickwrap cases from Specht through Hoffman makes clear, this entails, at a minimum, notifying users of the existence of such terms on the site’s home page (for example, through a clearly marked link to such terms), in a manner that is conspicuous and easily viewed by site visitors.  Moreover, when goods and services are available for purchase, it is recommended that website owners require customers to affirmatively acknowledge their acceptance of applicable terms before a purchase is completed.


A pair of recent decisions in federal court in Arkansas confirms that nothing about the virtual world changes a core principle of contract formation—that there can be no valid contract without objective manifestation of assent.  The decisions both deal with the efforts of one repeat pro se plaintiff, David Stebbins, to impose upon large institutions binding agreements to arbitrate via email.  These two decisions signal that courts will not relax traditional rules of contract formation merely because of the informality and relative ease of online communication.

The first decision, Stebbins v. Wal-Mart Stores Arkansas, LLC, No. 10-cv-3086, 2011 WL 1519390 (W.D. Ark. Apr. 14, 2011), relates to Stebbins’ interactions with Wal-Mart.  After applying unsuccessfully for a number of jobs at a local store, Stebbins sent an email to the company’s customer service department purporting to extend to Wal-Mart a formal offer to arbitrate any dispute between him and the company through an online arbitration service.  In the email, Stebbins explained that contact by anyone from Wal-Mart, in any form at all, would constitute agreement to be bound by the terms of the email, including submitting to arbitration.  The company responded with generic emails directing Stebbins to another department.  Meanwhile Stebbins, believing that Wal-Mart had accepted his email offer by allowing him to pay by check for a gallon of milk, registered with the online arbitration service, which emailed Wal-Mart that Stebbins intended to arbitrate an employment dispute with the company.  When Wal-Mart did not accept the invitation to arbitrate within 24 hours (a condition imposed by Stebbins under a “forfeit victory” clause in the purported contract), Stebbins claimed that he was entitled to a default arbitration award of over $600 billion, regardless of the merits of the dispute.

In another dispute before the Arkansas federal court, Stebbins v. University of Arkansas, No. 10-cv-5125 (W.D. Ark. May 19, 2011), Stebbins sought a similar agreement with the University of Arkansas relating to his unsuccessful attempts to re-enroll at the university following a suspension in 2007.  Stebbins already had a discrimination suit pending against the university for allegedly failing to accommodate his mental health disability.  While motions to dismiss were being considered, Stebbins emailed the General Counsel’s office a link to a YouTube video containing an offer to arbitrate all legal disputes and specifying that he would deem the offer accepted if the university communicated with Stebbins in any way, allowed him to communicate with university officials, or permitted him on campus.  Stebbins claimed that counsel for the university accepted his offer by fielding a follow-up telephone call and that, by failing to accept the invitation to arbitrate within 24 hours, the university lost the dispute under a “forfeit victory” clause.  Stebbins sought over $50 million and re-enrollment in the university.

Acting pro se, Stebbins moved to confirm both arbitration “awards” in separate actions in Arkansas federal court, a venue in which he had similar actions pending against his landlord and an online arbitration service.  In both cases, the court shut Stebbins down.  In the Wal-Mart case, the court explained that Stebbins could not rely on the concept of unilateral contract—in which a party accepts an offer to contract by performance instead of by express agreement—to prove the existence of a contract.  Stebbins’ emails were merely “self-serving documents that did not form the basis for any conduct or performance on Wal-Mart’s part,” and, indeed, “Wal-Mart performed no act.”

Similarly, in the University of Arkansas case, the court declined to accept Stebbins’ “novel proposition that one party can force a contract on another by sending an offer to contract, and stating therein that conduct entirely unrelated to a showing of agreement to be bound will constitute acceptance.” Distinguishing authority about the enforceability of clickwrap and browsewrap agreements, the court found that Stebbins had failed to demonstrate that the university showed any objective manifestation of assent to the formation of a contract.  The court explained that, if acceptance could be manifested in the ways Stebbins suggested, a contract might be formed if a university employee greeted Stebbins in a grocery store.  But “[t]his, of course, is not how contracts are formed, even on the Internet.”

This pair of cases posed relatively straightforward questions for the courts.  The attempts at contract formation were so one-sided, and the terms of the purported awards so outlandish, that the courts seemed to have little trouble dismissing the claims.  But the cases reinforce a basic notion that might provide comfort to institutions in closer cases—objective manifestation of assent is required no matter what method of communication is used to form a contract.