To promote securities market fairness, Regulation Fair Disclosure, or Reg FD, and Section 13(a) of the Securities Exchange Act of 1934 prohibit public companies and persons acting on their behalf from selectively disclosing material, nonpublic information to certain securities market professionals or shareholders where it is reasonably foreseeable that they will trade on that information, before that information is made available to the general public.  Under Reg FD, such information must be publicly filed with the SEC in a Current Report on a Form 8-K unless disclosed in a way that is “reasonably designed to provide broad, non-exclusionary distribution of the information to the public.”

To address the evolving landscape of electronic communications and issuers’ increasing use of websites to publish information, in 2008 the SEC released guidance that Reg FD applies to disclosures made through issuer websites, “push” technologies that allow automatic notification, such as email alerts and RSS feeds, and blogs.  The lynchpin of the analysis of whether disclosures made through such nontraditional electronic means comply with Reg FD, the SEC stated, is whether the issuer has made investors, the market, and the general public aware of the channels it plans to use to disclose material, nonpublic information so that they have the opportunity to place themselves in the loop.

In 2012, the applicability of Reg FD and the SEC’s guidelines to social media posting, a method of disclosure not specifically contemplated by the SEC’s 2008 guidance, was tested.  Reed Hastings, the CEO of Netflix, published breaking news on his personal Facebook page:  Netflix monthly viewing had exceeded one billion hours for the first time.  Netflix did not file a Form 8-K with the SEC, issue a press release, or otherwise publicly announce the news.  Although Facebook members could subscribe to Mr. Hastings’s Facebook page, and 200,000 had at the time of the post, neither Mr. Hastings nor Netflix had previously used it to break material news or indicated to the investing public that it might be used as a communication channel for such information.   This prompted an SEC investigation that culminated in a decision on April 2, 2013 not to pursue an enforcement action but to use the opportunity to clarify that Reg FD and the SEC’s 2008 guidelines and framework for analysis apply to social media.

“One set of shareholders should not be able to get a jump on other shareholders just because the company is selectively disclosing important information,” George S. Canellos, the SEC’s acting enforcement chief, stated. “Most social media are perfectly suitable methods for communicating with investors, but not if the access is restricted or if investors don’t know that’s where they need to turn to get the latest news.”

In preparing to publish company information through social media, a public company should consider whether Reg FD applies to that disclosure and, if so, whether such disclosure is compliant.  The first step of the analysis is to determine whether any potential recipient of the social media post is a shareholder, securities professional, or other type of person enumerated in Reg FD.  Due to the open-ended accessibility of social media, this is likely to be the case.  If so, and the information is material and nonpublic, the information must either be simultaneously filed in a Form 8-K or disclosed in a manner “reasonably designed to provide broad, non-exclusionary distribution of the information to the public.”  The key to compliance without filing a Form 8-K, the SEC has indicated, is to take proper steps to ensure that the public is on notice as to each outlet for material, nonpublic information about the issuer and that access to those outlets is available to those who want it.  Proper notice can be given by including in periodic reports filed with the SEC and press releases information on the specific social media channels a company intends to use to disseminate material, nonpublic information and maintaining consistency in practice.

While Netflix and Mr. Hastings received a free pass on Mr. Hastings’s Facebook post, the SEC’s message is clear:  Execs, don’t try this at home.  “Although every case must be evaluated on its own facts, disclosure of material, nonpublic information on the personal social media site of an individual corporate officer, without advance notice to investors that the site may be used for this purpose, is unlikely to qualify as a method ‘reasonably designed to provide broad, non-exclusionary distribution of the information to the public’ within the meaning of Regulation FD,” the SEC stated in its Report of Investigation.  The safest bet is still to file sensitive disclosures in a Form 8-K, but the SEC’s latest guidance has opened the door to the use of social media posting as an independently sufficient form of disclosure.

Now might be an opportune time for issuers to dust off their electronic communications policies, clarify social media usage guidelines for employees, and take steps to make the public aware of each specific channel through which they expect to disseminate material, nonpublic information.

UPDATE (April 11, 2013):

Taking cues from the SEC’s Report of Investigation, on April 10, 2013 Netflix publicly filed a Form 8-K with the SEC in which it stated that, in addition to traditional channels of dissemination, Netflix uses social media to communicate company information. “It is possible that the information we post on social media could be deemed to be material information,” Netflix signaled.  Netflix also updated the main page of its Investor Relations website to mirror its 8-K filing.

In its 8-K and on its Investor Relations website, Netflix encourages investors, the media and others interested in the issuer to stay tuned to various Netflix social media channels:  the Netflix U.S. & Canada Blog, The Netflix Tech Blog, the Netflix Facebook page, the Netflix Twitter feed and, most notably, Reed Hastings’s personal Facebook page, on which Mr. Hastings posted the message in 2012 that sparked the SEC’s investigation and which Netflix had not previously indicated could be a source of material, nonpublic information about the company.

Netflix’s official notices increase the likelihood that its future disclosures of material, nonpublic information through the social media channels it has identified pass muster under Reg FD.  As the SEC indicated in its 2008 guidance and the Report of Investigation, however, whether a particular disclosure complies with Reg FD’s requirement that material, nonpublic information be disseminated in ways “reasonably designed to provide broad, non-exclusionary distribution of the information to the public” will depend on a multifaceted fact-based analysis.

On its own, an issuer’s public identification of specific social media channels as potential sources of material, nonpublic company information does not necessarily ensure that any future disclosures of such information through those channels will be found compliant with Reg FD.  Therefore, one practical point for issuers to note is that, once a social media channel has been publicly designated, further steps should be taken to establish that channel as a recognized source of disclosure.  Such steps may include, for example, consistently using that channel to disclose material, non-public information on an equal basis with the issuer’s other channels of distribution.

However, under Copyright Act Section 501, a non-exclusive licensee may not bring an action for copyright infringement. See, e.g.,  HyperQuest, Inc. v. N’Site Solutions, Inc., 632 F.3d 377 (7th Cir. Jan. 19, 2011). Accordingly, a Web site provider that seeks to litigate based on an improper use of user-generated content may need more. They may in fact need an exclusive license or an actual transfer of ownership in the underlying copyright.  The question is, can one obtain an exclusive license or assignment of copyright through online terms of use?

Precisely this scenario is presented in a current dispute between a real estate multiple listing service and an online real estate information aggregator over the copying of photographs and listing information. The service provider is asserting claims with respect to user-generated content as a copyright owner, based upon an agreement presented by the provider to its users when they upload photographs to the provider’s database.

Metropolitan Regional Information Systems, Inc. (MRIS), the plaintiff, operates a multiple listing service that serves licensed real estate brokers and agents. Brokers and agents who enter into a subscriber agreement with MRIS can upload their listing information, including photographs of properties, to the MRIS site, and then display those and other listings on their own Web sites.

The defendant, American Home Realty, Inc. (AHR), is an online service provider that aggregates real property listing information on its www.neighborcity.com site. According to AHR, the information that it presents comes from a variety of public domain and other sources, including the MRIS online database.

In March 2012, MRIS brought suit against AHR in federal court, asserting that material from its online listings was included on the AHR site without authorization, and, therefore, that AHR infringed MRIS copyright rights in those materials. According to an affidavit submitted in support of its motion for a preliminary injunction, MRIS relies on the following provision in its agreement with its users as the basis for its claim of ownership of the copyright in uploaded photographs:

All images submitted to the MRIS Service become the exclusive property of Metropolitan Regional Information Systems, Inc. (MRIS). By submitting an image, you hereby irrevocably assign (and agree to assign) to MRIS, free and clear of any restrictions or encumbrances, all of your rights, title and interest in and to the image submitted. This assignment includes, without limitation, all worldwide copyrights in and to the image, and the right to sue for past and future infringement.

The court granted a preliminary injunction against AHR in August 2012. Metropolitan Regional Information Systems, Inc. v. American Home Realty Network, Inc., 2012 U.S. Dist. LEXIS 121352 (D. Md. Aug. 24, 2012). In November 2012, the court reiterated its ruling while narrowing the previous scope of the injunction to cover only photographs.

AHR argued before the District Court in August that the MRIS agreement does not comply with Section 204(a) of the Copyright Act, which provides that, except where copyright ownership is transferred by operation of law, the transfer “is not valid unless an instrument of conveyance, or a note or memorandum of the transfer, is in writing and signed by the owner of the rights conveyed or such owner’s duly authorized agent.” AHR argued that Section 204(a) has not been satisfied because, among other things, when a user uploads an image to the MRIS database, “no document is signed” and the assignor is not specified. Further, it is not clear whether the purported assignor has any rights to assign, and if so, what the basis is for those rights. See Reply Memorandum of Defendant American Home Realty Network, Inc.’s Rule 12(b) Motion to Dismiss, Metropolitan Regional Information Systems, Inc. v. American Home Realty Network, Inc., No. 12-cv-954(D. Md. filed June 15, 2012), p. 4.

In its August ruling, the district court rejected AHR’s argument, finding that the “MRIS TOU constitutes credible evidence that MRIS’s users intended to assign their copyrights to MRIS through the electronic submissions of photographs, which would satisfy the relevant provisions of” the federal Electronic Signatures in Global and National Commerce Act, 15 U.S.C. §§ 7001 et seq. (“E-SIGN”).

Following the district court’s August ruling granting the preliminary injunction, AHR filed an interlocutory appeal in the Fourth Circuit challenging the order, and that appeal currently is pending in the Fourth Circuit. As part of that challenge, AHR asserts that Copyright Section 204(a) requires “a signed writing on paper for a valid copyright assignment.”  Supplemental Opening Brief of Appellant, Metropolitan Regional Information Systems Inc. v. American Home Realty Network, Inc., No. 12-1202 (4th Cir. Dec. 7, 2012) (emphasis added). The assertion that a signed writing on paper is required is predicated on AHR’s analysis of Copyright Office regulations concerning the recordation of copyright assignments.

AHR further argues that federal E-SIGN does not validate the purportedly defective clickwrap assignments at issue, because the lower court failed to make any factual findings that MRIS’s clickwrap process satisfies E-SIGN’s formal requirements. Specifically, AHR asserts that the district court made no finding that there was any “electronic sound, symbol or process that was “attached to or logically associated with a contract” as required by E-SIGN.

We await the Fourth Circuit’s guidance on this issue.

Why is this an important issue? For many Web site owners, a transfer of copyright ownership is not necessary, as a broad license grant from the user is sufficient to provide the rights needed for business purposes.  However, as illustrated in the MRIS case, business-to-business sites are often engaged in struggles with competitive sites that scrape, spider or otherwise use content (including user-generated content) without permission.  To the extent a business-to-business site uses user-generated content, the assertion of a copyright claim against unauthorized users of that content can be a valuable tool in those struggles, and can be a powerful argument in any legal action that ensures.

It is worth noting that two forms of notice of the arbitration provision were actually alleged to have been provided to the consumer: the post-transaction e-mail, and a clickable link to the “Terms and Conditions” of the agreement that was presented at the time of enrollment in the disputed program. The efficacy of the second form of notice was not before the court, however, because that issue was deemed to have been waived at the lower court level. Thus, whether the presentation of the clickable link to the terms was sufficient notice of the arbitration provision was addressed with the comment that it “might have created a substantial question” as to whether the arbitration provision was enforceable.

(Trilegiant proffered screen shots purporting to depict a transaction confirmation page similar to that displayed to the plaintiffs, which the court made available on its Web site here, along with screen shots of the enrollment offer page purporting to be similar to the display of the clickable link to the plaintiffs, here and here.)

Data Pass Marketing

This case involves a “data pass” marketing program. These programs, also referred to as “Web loyalty” programs, typically are presented by third-party marketers to consumers at the conclusion of a transaction with an online retailer, in the form of a discount or cash back offer. If the consumer responds to the offer, whether by an affirmative “click” or by the provision of personal information, their payment data is provided directly to the third-party marketer by the online retailer who collected it in connection with the underlying transaction. The consumer is then enrolled in a program in which their payment card is charged a monthly fee.

As of 2010, the direct passing of payment data in this manner is prohibited by the Restore Online Confidence Act unless, among other things, the consumer’s “express informed consent” is obtained. The Act also imposes other requirements aimed at clearly differentiating these third-party programs from the underlying retail transaction in which they are typically presented. The transactions involved in this case took place in 2007 and 2009, before the Act was passed, and this case raises some of the same issues that are addressed in that legislation. (The Act is discussed further in this prior blog post.)

Agreement Now, Arbitration Provision Later

The plaintiffs in this case brought suit in the U.S. District Court for the District of Connecticut alleging that at the time they responded to the offer at issue, they did not realize they were enrolling in a fee-based program. In response to the consumers’ lawsuit, Trilegiant and its parent company responded with a motion to compel arbitration. They cited the arbitration provision included in the Terms and Conditions to which, they argued, the consumers had assented at the time they enrolled in the program. The arbitration provision was included in the post-transaction e-mail sent to the plaintiffs.

The district court held that the arbitration provision was unenforceable  and the Second Circuit upheld, holding that under either California or Connecticut law, the e-mail did not provide sufficient notice to the consumers to support the conclusion that they had assented to the provision.

The Law is Unsettled

Although courts have increasingly embraced “assent now, terms later” contracting, Judge Sack stated, judicial acceptance of this principle has resulted in the “conventional chronology of contract-making” becoming “unsettled.” Two analytical approaches are possible on the facts presented, the court said. The first possibility is that the agreement between the parties, including the terms and conditions including the arbitration provision, became effective only after the receipt of the e-mail by the plaintiffs and their failure to cancel their memberships. The second approach is that an agreement to pay a fee in return for program benefits was formed at the time of the consumers’ initial enrollment, with the additional provisions of the terms and conditions, including the arbitration provision, constituting proposals for amendment to the existing contract.

But the analytical approach did not ultimately matter, the court concluded, because under either analysis, the e-mailed terms, including the arbitration clause, were never accepted by the consumers.

Assent to Post-Transaction Terms May Be Based on Conduct, but Conduct Must Be Coupled with Knowledge

The court did not dispute that assent to contract terms may be manifested by conduct, including the acceptance of a benefit, citing Register.com v. Verio (2d Cir. 2004). But that act of assent the court stressed, must be coupled with actual or constructive knowledge of the terms to which assent is being given. While that knowledge may be actual or constructive, the court said, there was no actual knowledge in this case.

To the argument that the failure to cancel constituted assent, the court concluded that, where “the purported assent is largely passive,” the issue of contract formation turns on whether a “reasonably prudent offeree” would be on notice of the provision. In the absence of actual knowledge, the question becomes whether the consumers were on inquiry notice of the provision. Only then, the court reasoned, would their failure to cancel constitute “an objective manifestation of their assent” to arbitration.

Post-Transaction, E-Mailed Terms Are Not Expected by Consumers

On the issue of inquiry notice, the court commented: “We do not think that an unsolicited e-mail from an online consumer business puts recipients on inquiry notice of the terms enclosed in that email and those terms’ relationship to a service in which the recipients had already enrolled, and that a failure affirmatively to cancel the membership will, alone, constitute assent.”

The “touchstone” on this issue, the court said, is “whether reasonable people in the position of the parties would have known about the terms and the conduct that would be required to assent to them.” Those circumstances involve conspicuousness of the term, the course of dealing between the parties, and industry practices. On these last two factors, it is not surprising that the court found e-mailed notice to be lacking in this consumer transaction.

The court noted, that, unlike the parties in Register.com, there was no prior relationship between the parties, and there was no other means by which a reasonable person would understand that disputes would have to be resolved by arbitration.

Shrinkwrap Cases Are No Help Here

The court particularly rejected the argument that a finding of assent to the e-mailed terms, at least in a consumer transaction, was supported by the rulings in Hill v. Gateway 2000, Inc. (7th Cir. 1997) and ProCD, Inc. v. Zeidenberg (7th Cir. 1996). In those shrinkwrap cases, the court reasoned, notice of the terms was provided when the customer opened the packaging for the goods, as the terms were “necessarily included with the product,” and the consumer would understand that unless the goods are returned, the consumer takes the product subject to those terms.

In contrast, Judge Sack said, e-mail notice was “temporally and spatially decoupled from the plaintiffs’ enrollment in and use of [the service].” Thus, e-mail notice “lacks a critical element of shrinkwrap contracting – the connection of the terms to the goods (in this case the services) to which they apply.” Further, because of the ease with which the consumer was able to enroll in the program, the receipt of the e-mail was unlikely to “raise a red flag” to alert the consumer to the proffer of legally significant terms.

What the program provider did in this case, the court found, was to effectively obscure the terms and conditions and the passive manner in which they would be accepted, and “made it appear – falsely – that being a member imposed virtually no burdens on the consumer besides payment.”

Judge Sack gave a nod to the policy reasons supporting the endorsement of the agreement now – terms later model endorsed by the Seventh Circuit in the shrinkwrap cases, but concluded that no policy rationale supported the manner in which the provider presented the terms in this case. Requiring express acknowledgement of receipt of the terms was just one of the “plethora of other ways” in which the provider could have met the minimum requirements of notice. Such an express manifestation of assent could not be found in the auto-debiting of the consumer’s payment card, either, as the court deemed those payments “too passive” to manifest assent to be bound.

Browsewrap and Clickwrap Cases Are No Help Either

While the issue of notice based on the presentation of the hyperlink was deemed to have been waived, the court did address it in a lengthy footnote, finally concluding that it “fell outside the browsewrap and clickwrap categories.” The court distinguished the presentation of the hyperlink as follows:

The presentation of terms on the screens in the case before us falls outside both the clickwrap and browsewrap categories. Unlike the paradigmatic browsewrap agreement,  in this case there is some indication near the button that a user must “click” in order to subscribe to the service, that the service includes additional terms and that the user assents to these terms by clicking the button. In contrast to the typical clickwrap agreement, however, the button itself does not make explicit reference to these terms in asking the end-user whether he or she assents to them. It only suggests that a user can sign up for the benefits of the membership by clicking “Yes.”

Where Does This Leave Service Providers?

Perhaps the most telling aspect of the opinion is the court’s expressed view that the program provider in this case sought to “obscure” the terms and conditions and make it “falsely” appear that there were no burdens imposed upon the consumer aside from monthly payments. Given that perspective on the facts, it is perhaps not surprising that the court concluded that there was no assent.

And, the opinion may reflect a general judicial hesitation to enforce certain kinds of ancillary provisions in consumer contracts, and in particular, a hesitation to enforce arbitration provisions. There are plenty of examples of such opinions, several of which we have blogged in the past: Defontes v. Dell (R.I. 2009) and Hines v. Overstock.com, Inc. (E.D. N.Y. 2009).

More broadly applicable principles that may be gleaned from the ruling include a strong judicial skepticism on the efficacy of assent now – terms later contracting involving consumers where the transaction does not fall in the shrinkwrap category. The opinion suggests that in transactions where the delivery of the good or service is not closely tied to the presentation of the terms, unmistakable notice of them should be provided, whether by e-mail or otherwise. And consideration should be given to requiring a specific act of assent following the notice.

TripAdvisor.com is one of the top travel information sites (we can refer to Hitwise for that opinion); it refers to itself as providing the world’s “most trusted travel advice.” Since 2006, it has published the “Dirtiest Hotels” list.  Kenneth Seaton is the proprietor of the Grand Resort Hotel and Convention Center in Pigeon Forge, Tennessee, which had the unhappy distinction of being named the top “Dirtiest Hotel” in the 2011 TripAdvisor rankings. Seaton responded with an action for defamation and false light invasion of privacy.

Noting that the First Amendment is at least passively implicated in any defamation case, the court grounded its ruling in the fundamental distinction between actionable statements of fact, and constitutionally protected “pure opinion, hyperbole, or rhetorical exaggeration.” But an opinion implying an assertion of objective fact can fall outside that constitutional protection, the court noted.  Thus, to state a cause of action for defamation or false light under Tennessee law, the court said, the plaintiff must assert the publication of either “a false or misleading statement of fact, or a statement of opinion that implies having a basis in defamatory facts.”

Where is the line between an assertion of “actual, objectively verifiable facts” and mere opinion? It is, the court said, “whether a reasonable person could understand the language in question as an assertion of fact, or, on the other hand, regard the language as merely hyperbolic opinion or rhetorical exaggeration.” In the case of the Dirtiest Hotels list, the court found, it “is clearly unverifiable rhetorical hyperbole” that reflects nothing more than the opinions of TripAdvisor’s millions of users. In a passage that is likely to be widely quoted in future “top ten” defamation cases, especially those involving consumer reviews, the court said:

TripAdvisor’s list is of the genre of hyperbole that is omnipresent. From law schools to restaurants, from judges to hospitals, everything is ranked, graded, ordered and critiqued. Undoubtedly, some will accept the array of “Best” and “Worst” rankings as impenetrable maxims. Certainly, some attempt to obfuscate the distinction between fact and opinion as part of their course of business. For those that read “eat here,” “sleep there” or “go to this law school” and are unable to distinguish measured analysis of objective facts from sensational “carnival barking,” compliance will be both steadfast and assured. Nevertheless, the standard, fortunately, is what a “reasonable person” would believe. A reasonable person would not confuse a ranking system, which uses consumer reviews as its litmus, for an objective assertion of fact; the reasonable person, in other words, knows the difference between a statement that is “inherently subjective” and one that is “objectively verifiable.”

Accordingly, the court found that the “Dirtiest Hotels” list was not defamatory, and declined to address the direct First Amendment challenge posed by Trip Advisor.

An Important Line of Defense for User Review Sites

This is an important opinion for sites that feature user reviews and then utilize those reviews to generate editorial content. Although Section 230 of the Communications Decency Act protects such sites from liability for the content of the user reviews themselves, the availability of that protection can become tenuous when the site can be regarded as an “information content provider” with respect to claimed defamatory or otherwise harmful assertions.

What this ruling establishes is the best defense ever for an online consumer review site: there was no actionable defamation from which the provider needs protection.

Although decided under Michigan law, the ruling construes a provision of the Uniform Electronic Transactions Act. The UETA has been enacted in every state except New York, Washington and Illinois. At issue was the applicability Section 9 of the UETA, “Attribution and Effect of Electronic Record and Electronic Signature,” codified in Michigan as M.C.L. § 450.839:

(1) An electronic record or electronic signature is attributable to a person if it is the act of the person. The act of the person may be shown in any manner, including a showing of the efficacy of any security procedure applied to determine the person to which the electronic record or electronic signature was attributable.

(2) The effect of an electronic record or electronic signature attributed to a person under subsection (1) is determined from the context and surrounding circumstances at the time of its creation, execution, or adoption, including any agreements of the parties, and otherwise as provided by law.

In Zulkiewski v. American General Life Insurance Co., 2012 Mich. App. LEXIS 1086 (Mich. Ct. App. June 12, 2012) (unpubished), the former beneficiaries of the life insurance policy (the decedent’s parents) contested the change of beneficiary in favor of the decedent’s new spouse. The record established that the some months prior to the decedent’s suicide, someone purporting to be the decedent established an online account on the insurance company’s site, having supplied (as was required) the decedent’s policy number, social security number, mother’s maiden name, and an e-mail address. A confirmatory e-mail was generated by the insurance company, with instructions to contact the company if the insured did not intend to create an online account. A few minutes later, the account was used to execute the online change of beneficiary request in favor of the decedent’s spouse. The insurance company sent an e-mail message confirming that change, along with a letter confirmation sent by regular mail.

Following the decedent’s suicide, the former beneficiaries challenged the payout of the policy to the spouse, asserting that there was no proof that the decedent had requested the beneficiary change. They pointed out that while the UETA provides that an electronic record or signature cannot be denied effect simply because it is in electronic form, the electronic signature must be executed or adopted by the signatory as the embodiment of his or her intent. See M.C.L. § 450.832(h) (UETA Section 2(8)) (definition of an electronic signature). Referencing M.C.L. § 450.839(1) above, they argued that the insurance company was required to show the efficacy of its security procedures in order to establish the authenticity of the decedent’s electronic signature.

The Michigan appeals court disagreed, finding that the UETA does not require a showing of the efficacy of security procedures in order to establish the validity of an electronic signature. The court noted that M.C.L. § 450.839(1) provides that attribution of an electronic signature may be shown “in any manner,” with the efficacy of security procedures being one such showing. The court noted, among other things, that the person who made the online change of beneficiary had numerous items of the policyholder’s personal information; that both e-mail and postal mail notifications were sent upon the establishment of the online account and upon the change of beneficiary; that the policyholder was “highly computer-literate,” and that the policyholder had made the spouse the beneficiary on other insurance policies as well. The court also noted the spouse’s denial of any involvement in the making of the disputed change of beneficiary.

The court concluded that the insurance company had met its burden under Michigan law to establish its right to summary disposition of the issue of the validity of the signature. The court characterized the former beneficiaries’ argument as conjecture, and lacking support of any substantively admissible evidence.

While the ruling is grounded to some extent in Michigan procedural law concerning the evidentiary burden in a motion for summary disposition, it is a useful precedent construing the UETA. M.C.L. 450.836(c) (Section 6(3) of the Act), provides that the Act shall be construed to “[e]ffectuate its general purpose to make uniform the law with respect to electronic transactions among the states.”

The appeals court held that Section 230 is inapplicable to the plaintiff’s negligent supervision claim because an employer’s duty to supervise its employee “is distinct from any conduct like editing, monitoring or removing offensive content published on the Internet.” Viewed narrowly, as an opinion about employer liability for employee conduct, this rationale may make sense.

However, the Lansing court does not limit its analysis of CDA Section 230 to the employer-employee context. In statements that speak to the scope of CDA Section 230 generally, the Illinois court rejects a long line of federal court rulings that have construed the statute very broadly, in favor of the narrow view of the statute espoused by Judge Easterbrook of the Seventh Circuit in Doe v. GTE (7th Cir. 2003). Under his view, the statute would protect service providers from liability for decisions to remove third-party content, but would leave open the possibility of liability claims under state law for failure to remove such content.

The appeals court in Lansing expresses concern that the application of CDA Section 230 on the facts presented might produce what it views as an anomalous result, if the employer could potentially be held liable for the employee’s alleged threats made on the telephone, but not for threats made by e-mail or text message.  Under CDA Section 230, however, the law is different online, and the statute’s application often results in inconsistent results between on-line and non-online scenarios. These results may sometimes be difficult to accept, precisely because they are different from the results that might obtain cases involving offline communications. CDA Section 230 deserves its reputation as the law that judges love to hate, but most judges are able to set aside their opinions on the wisdom of the Congressional policy it expresses, and reject attempts to narrow the protection it affords online service providers.

The “broad immunity” that Judge Easterbrook and the Lansing court question is a principle that online service providers have relied upon for years in structuring and conducting their online enterprises. A contraction of the statute in the manner that they suggest would create a very different business environment for online service providers that deal with user-generated content.

If this general strategy sounds familiar, perhaps you have read our recent post on the Oracle v. Google dispute over Google’s use of Oracle’s Java technology in the Android operating system. In that case, the court ruled that Google had done it right, and that the rules and functionality of the Java technology that Google copied were not subject to copyright.

But in Tetris Holding, LLC v. Xio Interactive, Inc., 2012 U.S. Dist. LEXIS 74463 (D.N.J. May 30, 2012), Judge Freda Wolfson ruled that Xio, Ms. Golden’s development company, got it wrong. By wholesale copying not only the rules and functionality of the original Tetris game but also its copyrightable expression, Xio’s“Mino” app crossed the line into copyright infringement.

The key to the ruling in Tetris Holding v. Xio is the court’s deconstruction of the elements of the Tetris game into protectable and non-protectable elements. In doing so, Judge Wolfson referenced the same universe of software copyright rulings relied upon by Judge Alsup in Oracle v. Google.

Based upon these precedents, Xio argued in its moving papers that: “where a feature of a videogame is dictated by functional considerations, regardless of whether there may be a number of different ways to implement that feature’s functionality, copyright does not protect that feature.” Judge Wolfson rejected that theory as incorrect as a matter of law and logic:

“If an expressive feature is dictated by functional considerations then there cannot be a number of ways to implement it. Rather, one’s original expression is protected by copyright—even if that expression concerns an idea, rule, function, or something similar—unless it is so inseparable from the underlying idea that there are no or very few other ways of expressing it.”

The court also rejected Xio’s argument that the Mino game’s visual elements fell under the doctrines of merger and scenes a faire. The court pointed to testimony of Xio’s own expert, that, due to the fanciful nature of the game, there are an almost unlimited number of ways to design the game’s boards, pieces, movement and rotation.

Summary judgment was granted in favor of Tetris on its copyright infringement and trade dress claims.

And so, for Mino, it’s game over.

As appropriate to a dispute that presiding judge William Alsup referred to as “the World Series of technology litigation,” the API ruling is world-class.  The opinion will quickly make its way into technology law textbooks, not only for the clarity of the court’s discussion of how the Java programming language functions, but for its mini-treatise on software copyright law.

Whether the ruling will survive the already announced appeal is another issue, of course. But it should be noted that the due to the patent issues in the case, the appeal will be heard in the Federal Circuit, where familiarity with complex technology issues may, in general, be assumed.

That Judge Alsup’s dissection of the Java programming environment would be both extremely granular and technologically astute was no surprise to anyone who followed the trial. The court’s questioning of key witnesses, coupled with a series of technically detailed questions to the litigants and numerous rounds of briefs on software copyright issues, showed a judge who was deep into the details of software programming.

But the dramatic reveal of Judge Alsup’s technology chops came well into the trial, in an off-hand comment. In the course of debunking Oracle’s argument that Google had gained an advantage in copying a small amount of code into the Android operating system (A function called “rangeCheck,” not the APIs that are central to the dispute), Judge Alsup commented: “I have done, and still do, a significant amount of programming in other languages. I’ve written blocks of code like rangeCheck a hundred times before. I could do it, you could do it.”

A judge who codes.  Not your ordinary federal district court judge.

Here are some of the high points of the opinion.

• As noted above, the court’s explanation of the functioning of the Java computer programming language was a tour de force. If you don’t know what an API is when you start reading the opinion, you’ll know what it is when you’re done. One former computer science instructor, commenting on the ruling, remarked that Judge Alsup had “explained the technical basis of his ruling more clearly than many a CompSci professor could do.”
• The technical explanation was a necessary precedent to the deconstruction of the Java API into individual, potentially protectable elements, each of which required separate analysis under copyright law. Four separate elements were identified and analyzed: The names of the methods, classes and packages in the API; the structure, sequence and organization of these elements; the method specifications; and the implementation of the methods. This iteration of the Second Circuit’s “abstraction, filtration comparison” test is likely to be the roadmap for analyzing similar software copyright issues in future cases.
• Oracle’s copyright claim in the names of the methods, classes and packages failed under the merger doctrine and the short phrases rule.  Similarly, the copyright claim in the method specifications failed because the specifications in the Android API were required by the rules of the Java programming language to be identical to those in the Java API in order for the Android API to function with the Java language. Accordingly, the court held that the merger doctrine barred Oracle’s copyright claim.
• Analysis of the tension between copyright law and patent law was central to the claims in the structure, sequence and organization of the API. The court relied on Copyright Act Section 102(b), which precludes copyright protection of an idea, procedure, process, system, method of operation or concept. The court held that allowing Oracle to claim copyright in functional aspects of the API, even though they were the product of a creative endeavor, would bypass the 20 year limitation on patent claims.
• The court cited commentary suggesting that broader patent protection for software has lessened the need for copyright protection for potentially patentable elements such as the structure, sequence and organization of programs. (One may wonder how that comment will go down in the Federal Circuit, which has just had the Supreme Court remand back to it a software patent case for reconsideration of its breadth.)
• The court recognized Oracle’s claim in the implementation of the methods embodied in the API. However, it was conceded that these elements were not copied, that Google had written its own implementations of the Java API.
• The court did not mention the recent ruling of the European Court of Justice in SAS Institute Inc. v World Programming Ltd. Case C-406/10 (May 2, 2012) (See prior blog post.)  This was surprising, as the court requested comment on the ruling, and some reference to its relevance, or lack of relevance, to the Oracle v. Google dispute might have been expected.

Note also that in a separate ruling, Judge Alsup rejected Google’s equitable defenses of implied license and waiver. The court found that the failure of Sun (Oracle’s predecessor in interest) to take action against Google when it learned of its use of the Java technology was mere inaction, not the affirmative grant of permission necessary for an implied license, nor the overt act indicating an intention to abandon a known right necessary to establish a waiver. The court reserved decision on the defenses of equitable estoppel and laches in the event of a remand.

Given the certain appeal, this is not the last we will hear of this important case. Whether the appellate judges agree with Judge Alsup or not, he has laid out the issues with a remarkable clarity that will aid them in their review.

The dispute in SAS Institute Inc. v World Programming Ltd. Case C-406/10 (May 2, 2012) involves the SAS Language that was developed by SAS for scripts, or application programs, that are run in conjunction with SAS’s Base SAS database program to extract data. Using commercially available SAS products, World Programming developed alternative software capable of running scripts originally written in the SAS language.

SAS brought an action against World Programming in the UK, alleging that World Programming infringed its copyrights by copying SAS manuals and other components of the SAS software. Central to the case is certain provisions of the European software copyright directive, Council Directive 91/250/EEC of 14 May 1991 on the legal protection of computer programs (OJ 1991 L 122, p. 42). The UK High Court referred issues concerning the scope of that directive to the EU Court of Justice for resolution.

The software copyright directive extends to “expression in any form of a computer program,” but it excludes “ideas and principles which underlie any element of a computer program, including those which underlie its interfaces.” The Court of Justice concluded that World Programming’s activities for the most part fall on the non-copyright side of this equation, because “neither the functionality of a computer program nor the programming language and the format of data files used in a computer program in order to exploit certain of its functions constitute a form of expression of that program.”

The Court of Justice also ruled that a licensee of a computer program may use reverse engineering techniques to determine the program’s principles and functionality. The court referred the question of infringement resulting from the copying of the SAS manuals back to the UK court for decision under UK law, ruling that under the software copyright directive such copying “is capable of constituting an infringement.”

The entire case will now be returned to the UK High Court for decision. This ruling by the EU Court of Justice is interpretive only; it is up to the national court to implement it.

The SAS Institute v. Word Programming dispute is well-known to the parties in Oracle, Inc. v. Google, Inc., who have disagreed on its potential relevance even before the court’s May 3 order asking them for comment. See, e.g., Google, Inc.’s Copyright Liability Trial Brief (Apr. 12, 2012) and Oracle’s Brief Regarding Copyright Issues (Apr. 12, 2012).

We should learn very soon whether the result in Oracle, Inc. v. Google, Inc., will be consistent with this latest development in EU intellectual property law.