Google vs. Bing: An IP Law Perspective

Last week, Google accused Bing, Microsoft’s search engine, of copying its search results. Since Google first made this accusation, Bing has vigorously defended itself and the spat has mushroomed into an all-out public feud.

This dispute between two market giants fascinates me. It raises all sorts of interesting questions: What is ethically appropriate for a search engine? Is there an objectively ideal set of search results or are searches necessarily shaped by their designers? And, most importantly, are there any viable intellectual property claims against Bing? I will focus primarily on this last question in this blog post.

First, a word regarding Bing. As most people know, Microsoft is the company behind the Bing search engine. Throughout this post, I will refer to the organization within Microsoft that designs and maintains the Bing search engine as “Bing.” Sometimes I may refer to Microsoft instead; but where I choose one name over the other, I am not attempting to draw any meaningful distinction. For the purposes of this post, the two are interchangeable.

Some Background. In May of last year, Google started noticing that Bing was suddenly much better at returning results for unusual misspellings, often returning the same set of results that Google might return. Danny Sullivan, over at Search Engine Land, provides a few examples: “torsoraphy” instead of “tarsorrhaphy” or “bombilete” instead of “bombilate”. Because Google has worked hard on what it characterizes as a cutting-edge spelling correction system, it took immediate notice of this apparent copying by Bing and started watching Bing more closely. In addition to correcting misspellings, Google also prides itself on its fairly robust “long-tail” searches—searches for obscure terms or phrases where there are few results. It seemed to Google that Bing was leveraging Google’s hard work in this area and getting a free ride.

Then, in October of last year, Google saw a “marked rise” in similarities between Google search results and Bing search results, including the result placed in the top spot for any given search. Because the results were so strikingly similar, Google decided to initiate a sting operation. You can read about exactly how Google’s sting operation worked at Search Engine Land. Suffice it to say that Google introduced artificial or synthetic search results for truly random words like “hiybbprqag” and “indoswiftjobinproduction”. At the start of the Google sting, these words produced no results on either search engine. Once Google hard-coded the fake results pages for these queries and sent its engineers home to run seed searches using Internet Explorer and the Bing toolbar, it wasn’t long before these results began to show up at bing.com, in roughly 7–9 percent of searches. Looks like a clear-cut case of copying right?

IE and the Bing Toolbar. Google suspected that Microsoft was doing its copying primarily by collecting information about Web surfing habits from users of two of its products: Internet Explorer (IE) and the Bing toolbar. IE offers users a “suggested sites” feature that, when enabled, transmits clickstream data (lists of sites users visit and searches performed at those sites) back to Microsoft. Similarly, Microsoft’s Bing toolbar, which users can install in IE or Firefox, collects and sends clickstream data back to Microsoft. Microsoft does notify users of its collection of such data in its install dialog, which links to its terms and conditions and privacy policy, and permits users to opt-out of such collection. Pundits and lawyers disagree about whether Microsoft does enough to notify users about precisely what it’s doing with the data. I won’t examine that issue in this post.

Microsoft admits to using such clickstream data as input to its search engine algorithms. It is, in essence, crowd sourcing its search results by watching what users search for and click on at websites across the Internet. However, Bing is careful to point out that the clickstream data is just “one small piece” of over 1,000 different signals that feed into the Bing ranking algorithm. Furthermore, Bing isn’t targeting Google per se; it’s just that, because of its relative popularity, Google makes up a significant portion of the Bing clickstream.

To complicate matters further, it is possible that Google is using clickstream data collected via its own Google toolbar to improve its own searches. Google has so far denied this but there is some evidence to the contrary.

Aether. What happened in October to cause a noticeable change in Bing’s search results? Bing rolled out a new (and presumably better) ranking algorithm (something code-named “Aether”). According to Bing, that explains the sudden shift Google witnessed.

So far, Google has not made any intellectual property infringement or misappropriation claims against Bing. And, as you’ll see below, it is doubtful that they will. Google’s argument in their very public dispute with Microsoft over Bing’s reliance (in part) on Google results as input for their search algorithm is that Bing is behaving unfairly, and possibly unethically in its use of Google results.

The refrain from Google since this story broke has been, “you’re copying our hard-earned search results!” And Bing’s has been, “Google’s results are 1 of 1,000 data points Bing uses to improve its search results.” As far as I can tell, they’re both right. So the question is, does this skirmish need a referee? Should the law, the government, step in to decide a winner and a loser in at least this battle of the larger search engine wars? The answer is probably no.

Within the legal regime that we have here in the US, in light of the intellectual property laws and internet/cyber laws that might apply, this type of competitive behavior is permitted. In short, Bing is not copying anything that’s protected by copyright. Whether Bing infringes any patents in its gathering clickstream data is conceptually unrelated to the question of improper copying here; either Bing infringes a patent covering clickstream analysis or not. It would not be Google-specific. Perhaps there is an open question around whether Microsoft is using improper means to discover a trade secret. But even that is probably a stretch, as we’ll see below. Regarding trademarks, copying search results does not raise a trademark issue beyond those raised by generating search results as a general matter.

Let’s explore the relevant intellectual property doctrines in more detail.

Trade Secrets. Google and Microsoft both rely on trade secret protection (and, to some extent, copyright) for their search algorithms. As with the formula for Coca-Cola or the manufacturing process of K-2 skis, when the value of an invention to its owner is directly proportional to its secrecy, then it makes perfect sense to choose trade secret protection over patent protection. Also, patents only provide protection for about 20 years while trade secrets are protected for as long as they remain secret.

As I have explained in a previous post, a trade secret is information not generally or publicly known for which reasonable precautions have been taken to protect it from public disclosure. If someone uses improper or wrongful means to discover a trade secret, that person can be liable for trade secret misappropriation. Improper or wrongful conduct is that which falls below generally accepted standards of commercial morality and reasonable conduct. Reverse engineering is not typically considered improper in the absence of a binding non-disclosure agreement (NDA).

Given this definition, based on what we know about Microsoft’s actions in using Google’s results in its own search results, the individual search results that Microsoft has collected are likely not trade secrets. They are public; anyone can obtain them by searching Google. And Microsoft’s collection via its user base is authorized by its end-user agreements. Perhaps an enterprising lawyer could argue successfully that Microsoft’s agreements with users of IE or the Bing toolbar are improper or wrongful. After all, Microsoft could do a better job disclosing to users that it may use the user data it collects to improve the Bing search engine. But supporting a claim of misappropriation based on this seems like a stretch, especially given the public nature of what’s being “copied.”

Copyright. What about copyright protection for the search results gleaned from Google by Bing via the IE Suggested Sites feature or the Bing Toolbar? There are several impediments to granting Google copyright protection for individual or collected search results.

Search results are merely a collection of facts about other websites, generated by a machine, whose order is based on an algorithm attempting to offer users useful information. Copyright does not protect facts and requires that copyrightable subject matter be (at least minimally) creative and original, not utilitarian. Of course there is more to it than that but that’s a layman’s summary of the relevant threshold to copyrightability for this kind of expression. Copyright lawyers will recognize that search results are more akin to the unprotectable “sweat of the brow” exemplified by the US Supreme Court in Feist. At most, Google could claim infringement of its copyright for the entirety of its search results as a compilation, or the look and feel of its results pages. But that would require that Bing (substantially) copy the results for a particular search term en masse, or the specific look and feel of the Google results pages.

Unfair Competition? Perhaps there is a viable state-law claim for unfair competition, maybe some sort of legally-actionable, deceptive practices on the part of either Bing or Google. Given the facts made public so far, however, it is not clear that either party is engaged in any behavior that would qualify under the various state laws in this area, especially once we have ruled out claims related to trade secret misappropriation and trademark infringement.

Conclusion. There do not appear to be any agreed-upon standards or industry norms for how a search engine company is to gather data to improve its search algorithm. In my opinion, Bing’s crowd-sourcing approach is not facially unethical and probably not illegal as a general matter. As suggested by Search Engine Land and others, this is most likely a series of PR maneuvers by Google and Bing that will eventually dissipate.

Bing characterizes Google’s sting operation as a “spy-novelesque stunt to generate extreme outliers in tail query ranking,” a “honeypot attack,” and even “click fraud.” To the engineers at Bing, clickstream data is fair game and Google is abusing its market power in search to stifle or reframe what is otherwise fair competition. In addition, according to Microsoft, Google has chosen the timing of its allegations of copying to deflect attention from the recent media and blogosphere stories about what many are calling a decrease in search results quality at Google.

Is it possible that all search engines, if given enough time and data points, would eventually converge on an ideal set of search results? Is that what’s happening here? Or is search inherently subjective, are search results always the result of an idiosyncratic set of decisions made by the designers of a particular search engine? If it’s the former then Google will increasingly find other engines returning similar results. However, it does appear that search engines are (and should be) a reflection of the priorities chosen by the designers, what Sullivan calls a search engine’s “search voice.” For Google this might mean weight X on inbound links and weight Y on social graph. For Bing the priorities might be weight X on actual user behavior (clickstream) and weight Y on inbound links. Blekko might emphasize more of a curated approach (as it appears to be doing).

Since a diversity in “search voice” ultimately benefits consumers and gives a search engine its competitive edge, Bing should naturally be incentivized to place less emphasis on the Google clickstream. In any case, it appears that Google will have to give Bing the leeway to innovate with the clickstream and its crowd-sourced approach to search. In the end, Blekko may have the most to gain from this squabble.

The Removal of the VLC App from the App Store: A Lesson in Open-Source Licensing

A few months ago, I explained how the GPL conflicts with the Apple App Store terms. With the popular VLC app being pulled, I would like to take this opportunity to walk through the reasoning that likely led to Apple’s decision to pull the app. This is not based on any inside information I have from Apple, just my own reading of the publicly available information.

If there is one thing that readers should take away from this post (and my prior post), it’s that the GNU General Public License (GPL) and apps from the Apple App Store do not mix. To the extent that issues can be settled within open-source licensing jurisprudence, this is not a controversial conclusion; given the wording of the GPL, this is the logical legal conclusion.

Nevertheless, engineers and technology enthusiasts often misunderstand the purpose and effect of the GPL, thinking that it means an absence of copyright or some sort of relinquishing of control by the authors of the software.

Take, for example, this strident and indignant post over at The Unofficial Apple Weblog (TUAW). Chris Rawson, the author of the post, characterizes Rémi Denis-Courmont’s request of Apple to remove the VLC app as mere “antics” and “one man’s misguided crusade.” (A prior post at TUAW by a different author employs a more reasoned and sympathetic tone.) Another blogger called Mr. Denis-Courmont a “vicious Nokia employee.” In actuality, based on publicly available information, this is simply a case of a copyright owner asking Apple to honor his rights under copyright law. So what led to this tempest in a teapot?

VLC is a software media player that has a reputation as the most robust and versatile video playback software available on any platform, supporting just about any video codec you can think of. (Its potential codec-related patent issues is an interesting topic for another time.) Last year some developers got together and created a port of VLC for iOS. The app and its corresponding source code are still available for download, presumably for those who have jailbroken their iOS devices, and to comply with the terms of the GPL.

On October 25, 2010, one of the primary contributors to the VLC Project, Rémi Denis-Courmont, requested removal of the VLC app from the Apple App Store, explaining that the GPL is “contradicted by the products usage rules of the AppStore [sic].” Some time during the first week of January, 2011, Apple removed the VLC app. Then all hell broke loose in the blogosphere.

According to the license notices on the software, VLC is available under the GPLv2 or “(at your option) any later version.” It appears that the VLC app was licensed pursuant to GPLv2. As you may recall, the App Store terms impose a number of usage rules, two of which conflict with Section 6 of the GPLv2. Because of this conflict, Apple and the VLC app developers have not satisfied the conditions of the GPL and are therefore ostensibly infringing certain copyrights in the software.

Regarding Denis-Courmont’s standing to file such a request for removal, he has that right as the copyright owner of the code he contributed to the VLC project. It does not appear that contributors to the VLC project are required to assign ownership of the copyright in the code they contribute to VideoLAN. So, barring some other assignment, Denis-Courmont still owns part of the VLC code. In response to the unwarranted outrage and vitriol directed his way, Rémi Denis-Courmont has defended his position at his VideoLAN blog.

Open-source licensing can be confusing and can often appear to go against the ideal of software freedom. I suppose another way to put it is that everyone has their own idea of freedom. The important thing to remember is that, when it comes to software licensing and copyright ownership generally, the copyright owner gets to define the scope of, and conditions to, use. And that is all that happened here.

UMG v. Augusto: Applying Vernor’s First Sale Test to Promotional CDs in the Ninth Circuit

On January 4th the Ninth Circuit Court of Appeals issued another decision on copyright’s first sale doctrine, applying its recent Vernor v. Autodesk holding to the sale of promotional music CDs and finding that Troy Augusto—the defendant in the case—had the right to sell compact discs (“CDs”) containing Universal Music Group (UMG) music. The case is UMG Recordings, Inc. v. Augusto.

Although the case hinged mainly on copyright’s first sale doctrine—17 U.S.C. §109—it did rest in part on the Unordered Merchandise Statute, which was enacted as part of the Postal Reorganization Act of 1970. However, that idiosyncratic wrinkle was not essential to the Court’s holding, as we will see.

Background. Like many record labels, UMG routinely sends promotional CDs to music critics, radio programmers and other members of the music industry. UMG typically sends these CDs via the United States Postal Service or UPS. Anyone who has purchased used CDs is familiar with the stamp often affixed to such CDs that reads, “Promotional Use Only—Not for Sale” or something similar. UMG relied on such statements in some cases, in others it included a lengthier notice:

This CD is the property of the record company and is licensed to the intended recipient for personal use only. Acceptance of this CD shall constitute an agreement to comply with the terms of the license. Resale or transfer of possession is not allowed and may be punishable under federal and state laws.

Through means unimportant to the reasoning in this case, Augusto acquired numerous UMG promotional CDs and sold them online, through eBay and other sites. He often acknowledged in his item descriptions that the CDs were promo copies. Consistent with the industry practice of leaving no stone unturned in an attempt to exercise complete control over music sales, UMG attempted to halt such online sales and eventually filed a complaint against Augusto in federal court for infringement of UMG’s exclusive right to distribute its CDs under the Copyright Act.

Augusto argued that he (1) was protected by the first sale doctrine and (2) that he had the right to do what he wanted with the CDs under the Unordered Merchandise Statute. UMG argued that the “promotion only” notices on the CDs rendered the transaction with recipients one of license rather than sale or transfer. Consequently, under Vernor, the first sale doctrine would not apply.

First Sale. Recall that the first sale doctrine as codified by §109 provides that “the owner of a particular copy or phonorecord lawfully made under [the Act], or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.” The central inquiry in a case like this is whether the defendant can be said to own the copies in question. Recall from Vernor that the Ninth Circuit considers three factors in making such a determination. Does the copyright owner:

1. Characterize the transaction as a license?
2. Significantly restrict the user’s ability to transfer the software?
3. Impose notable use restrictions?

If the answer to all three is yes then it is a license and not a sale. Here, because “the recipients were entitled to use or dispose of [the CDs] in any manner they saw fit, and UMG did not enter a license agreement for the CDs with the recipients,” the original recipients owned their copies, as did Augusto.

The Court placed great emphasis on the nature of UMG’s distribution. First, UMG makes no attempt to track individual CDs. Moreover, because they were mailed without the prior expressed request or consent of the recipients, the CDs are unordered merchandise under the Unordered Merchandise Statute §3009 and recipients can therefore do what they please with such “merchandise.” Apparently §3009 applies to shipments regardless of carrier.

Next, the Court makes clear that shipment of these CDs would be characterized as sales or gifts for the purpose of the first sale doctrine regardless of whether they qualify as unordered merchandise. Crucially, in contrast with a software license such as that found in Vernor, there was no affirmative acceptance on the part of the recipient to the “license terms” printed on the CDs. Without acceptance, there can be no license agreement re-characterizing the transaction as a license or imposing enforceable transfer or use restrictions.

Finally, the Court points out that recipients were also not required to return the CDs. Although this fact alone would not conclusively establish a sale in the Ninth Circuit, it is further evidence that the Vernor test was not satisfied.

A brief aside re FLOSS licensing: Are there consequences here for free/libre/open-source software licensing? After all, open-source licensing relies on one of two arguments to make open-source licenses enforceable: (1) use of the software and/or exercise of any rights under copyright serve as acceptance of the terms; and (2) the terms are conditions to the “naked” license granted thereunder and acceptance is irrelevant—either you honor the conditions or you infringe. Without the affirmative acceptance required by the UMG Court, open-source licensors would be forced to fall back on the latter argument, a position typically only advocated by the Free Software Foundation for its GNU licenses (e.g., the GPL and LGPL). Of course, because copying is necessary to distribute electronic copies, the first sale doctrine would only apply to software distributed on CD or other physical media. So the practical impact is likely minimal.

Conclusion. The Court was careful to distinguish the facts here—short notices stamped on music CDs—from the software licensing context of Vernor, stopping short of saying that its holding in Vernor should be limited to computer software. Although this decision does not substantially alter the first sale jurisprudential landscape, it should offer some comfort that Vernor will not be applied to traditional creative works (such as books, DVDs and music), at least where such works are distributed on physical media. To the extent that Vernor hammered another nail into the coffin of the first sale doctrine, this decision partially reanimates the corpse for non-software creative content distributed on physical media. Apologies for employing such a strained, TechCrunchesque metaphor.

The relative harm that sale of promotional CDs causes to the music industry must pale in comparison to that caused by internet file sharing and a failure to offer quality products in an increasingly competitive market for digital entertainment of all kinds. Regardless of the outcome of this case, it seems that UMG would have been better off focusing its substantial resources elsewhere.

The Ninth Circuit’s MDY v. Blizzard Decision: A Fresh Perspective on Software Licensing and the DMCA

The recent Ninth Circuit decision in MDY v. Blizzard touches on several important technology and internet law issues; namely, software licensing and anti-circumvention under the Digital Millenium Copyright Act of 1998 (DMCA). The case involves a very popular, massively-multiplayer-online-role-playing game (MMORPG) called World of Warcraft (WoW) and the facts feature a technological arms race in which the game makers successfully leveraged copyright and other claims after they effectively lost that arms race.

World of Warcraft (WoW) is an MMORPG created by Blizzard in which players role play in a fantasy world, an online world where they can interact with other players. With more than 10 million subscribers, WoW is one of the most popular games in computer game history. As you might guess, WoW is comprised of both client software and server software.

Glider. Starting in 2005, MDY Industries, LLC, a software company founded by Michael Donnelly, distributed “bot” software called Glider that automates play of WoW so that players can advance within the game (gain experience points, character levels, loot, etc.) without having to sit for hours in front of the computer. Glider achieves this by moving the mouse and pressing keys on behalf of the user. MDY argues that Glider does more than facilitate cheating; disabled players can use Glider to enjoy WoW and all players can derive more enjoyment from the game. In addition, since Glider does not modify or reproduce the local WoW client software—it runs as a separate process on the user’s computer—there can be no copyright infringement. However, as we will see below, according to the Ninth Circuit, the anti-circumvention provisions of the DMCA create at least one cause of action that is entirely divorced from a copyright claim.

Warden. As you can imagine, widespread use of Glider negatively impacts the player experience for other players. In fact, between 2004 and 2008, Blizzard received 465,000 complaints related to WoW bots. In response, Blizzard spends $940,000 annually on the bot issue.

In late 2005, Blizzard rolled out Warden, its technological response to Glider that was designed to detect an instance of Glider running on a player’s computer and to disable Glider. MDY then modified Glider to avoid detection by Warden and touted this new Glider feature on the Glider website. MDY even went so far as to concede on its website that Glider violated Blizzard’s Terms of Use (ToU) agreement.

At $15–$25 per copy of Glider, with WoW’s immense popularity, MDY had gross revenues of $3.5 million as of September, 2008 (having sold some 120,000 copies as of that date). With so much money at stake, after receiving a few nastygrams from Blizzard’s lawyers, Donnelly was motivated to file a claim for declaratory judgment in order to clear up any uncertainty. And, given MDY’s income, he could afford to litigate against Blizzard. In late 2006 MDY filed such a claim.

Agreements. There were two agreements at issue in this case: the Blizzard End User License Agreement (EULA) and Blizzard’s online Terms of Use (ToU). The EULA governs Blizzard customers’ use of the local WoW client software and the ToU governs users’ access to Blizzard’s server software and online service, in which WoW players connect to each other in the WoW virtual world.

District Court. The district court opinion was a resounding win for Blizzard. The lower court found MDY liable under DMCA §§ 1201(a)(2) and (b)(1), and it found Donnelly liable on all counts: secondary copyright infringement, violations of two of the anti-circumvention provisions of the DMCA, and tortious interference with contract. In this blog post, I will focus on the Ninth Circuit’s holding on the copyright infringement and DMCA claims, in that order.

Secondary Copyright Infringement

In order for MDY to be liable for secondary copyright infringement—either contributory or vicarious—there must first be direct copyright infringement. In other words, Glider users must be infringing Blizzard copyrights in order to establish MDY’s secondary liability.

In using the client software, a WoW player’s computer makes a copy of the client software in RAM. At least in some federal courts, this would be an infringement of Blizzard’s exclusive right to reproduce the software unless players “own” their copy of the client software or have a valid license from the copyright owner. Some courts have held that RAM/buffer copies are not infringing copies.

Do Players Own or License the Client Software? – The Essential Step Defense

If WoW players own their copy of the client software then Glider users have a legal right—under 17 U.S.C. §117—to make RAM copies and there can be no secondary copyright infringement. The Ninth Circuit most recently examined this issue in Vernor v. Autodesk, Inc., a Ninth Circuit decision from September 10th of this year. In Vernor, the Court held:

that a software user is a licensee rather than an owner of a copy where the copyright owner (1) specifies that the user is granted a license; (2) significantly restricts the user’s ability to transfer the software; and (3) imposes notable use restrictions.

In light of Vernor, because Blizzard reserves title in the client software, grants players a limited license, and imposes transfer restrictions, the Blizzard Court found that WoW players are mere licensees of the WoW client software, that they do not own their copies of the client software. Notably, the transfer “restrictions” in the EULA do permit transfer so long as players:

physically transfer[] the original media (CD-ROM or DVD), all original packaging, and all Manuals or other documentation distributed with the Game; provided, however, that [players] permanently delete all copies and installations of the Game in [player’s] possession or control, and that the recipient agrees to the terms of [the EULA].

Apparently, despite the relatively permissive nature of this provision, the Court deemed this a sufficient transfer restriction. Therefore, given the Ninth Circuit’s decisions on this issue, one could reasonably conclude that anything short of the right to transfer unconditionally would satisfy this prong of the Vernor test. Still, the transfer provision contained in the Blizzard EULA appears to describe exactly the sort of transfer rights the owner of a copy would have. Unfortunately, the Court does not offer any further explanation on this point.

Because Glider users could not take advantage of the “essential step defense” provided by 17 U.S.C. §117(a)(1), the Court next considered whether the language of the EULA precluded a claim for copyright infringement.

Conditions versus Covenants

To determine whether a license grant is conditional, and therefore leaves open a claim for copyright infringement where the conditions are not satisfied, a court looks to the language of the applicable agreement(s). Generally, where a copyright owner grants you a license and you violate the terms of the license agreement, it may only sue you for breach of contract. The license agreement is effectively a waiver of the right to sue for copyright infringement. However, properly drafted limitations or obligations in a license agreement can act as conditions to the grant of license, as opposed to contractual covenants (binding promises), thereby opening the door to a copyright infringement claim (in addition to the breach-of-contract claim).

What separates a condition from a covenant is a matter of state law and may vary from state to state. In general, and in Delaware (the governing law of the agreement here), a covenant is “a contractual promise, i.e., a manifestation of intention to act or refrain from acting in a particular way.” By contrast, a condition (or, more accurately, a “condition precedent”) is “an act or event that must occur before a duty to perform a promise arises.” For policy reasons, courts favor covenants; where there is ambiguity, a court will find a covenant.

The WoW EULA grants players a license to (a) install the client software on “one or more computers owned by [the user] or under [the user’s] legitimate control” and (b) use the client software in conjunction with the WoW service for “non-commercial entertainment purposes only.” The EULA (and the ToU) also contain certain express limitations, such as a prohibition on using “cheats, bots, ‘mods,’ and/or hacks, or any other third-party software designed to modify the World of Warcraft experience.” Applying the interpretive principles outlined above, it is fairly clear that the prohibitions against bots and unauthorized third-party software are covenants rather than conditions. Consequently, according to the Court, Glider users are not infringing Blizzard copyrights; they are simply breaching the terms of the EULA (and the ToU).

As an IP licensing attorney, what is interesting and somewhat perplexing about the Court’s analysis and conclusion here is that the license grant itself is expressly conditioned on the user’s compliance with the terms of the agreement, arguably creating a condition. So the holding negates one common license drafting heuristic, which says that one should always make the license grant subject to the licensee’s compliance with the terms of the agreement in order to make it conditional and preserve a copyright infringement claim. Although I am not certain that the WoW EULA I have reviewed is the one at issue in this case, it is from 2007-2008, making it highly likely to be the agreement at issue. And that EULA does contain relatively standard conditional prefatory language in the license grant section:

Subject to your agreement to and continuing compliance with this License Agreement, Blizzard hereby grants, and you hereby accept, a limited, non-exclusive license to (a) install the Game Client on one or more computers owned by you or under your legitimate control, and (b) use the Game Client in conjunction with the Service for your non-commercial entertainment purposes only.

In short, in the language of the Court, “for a licensee’s violation of a contract to constitute copyright infringement, there must be a nexus between the condition and the licensor’s exclusive rights of copyright.” Although those concerned that copyright law already grants copyright owners too much power are likely pleased with the Court’s conclusion on this point, licensors in the Ninth Circuit should take note.

Because there could be no direct copyright infringement, there could be no secondary copyright infringement on the part of MDY or Donnelly.

DMCA Claims

Resolution of Blizzard’s DMCA claims posed questions of first impression for the Ninth Circuit and, as the Court points out, raised issues yet unresolved at the U.S. Supreme Court. The two DMCA anti-circumvention claims are as follows:

• §1201(a)(1)(A): prohibits circumventing a technological measure that effectively controls access to a work protected under the Copyright Act
• §1201(b)(1): prohibits trafficking in technology that circumvents a technological measure that “effectively protects” a copyright owner’s right

The trial court considered these claims with respect to three WoW components:

• the game client software’s literal elements (the computer source code to the game)
• the game client software’s non-literal elements (the over 400,000 visual and audible components, “such as a visual image of a monster or its audible roar”)
• other dynamic non-literal elements, the combined overall experience of traveling through the game world, that is provided in part by Blizzard servers (in combination with the client software)

1201(a)(1)(A). Because players already have access to the client software, the primary question here was whether Warden effectively controlled access to WoW’s dynamic non-literal elements. The trial court found that MDY violated both 1201(a)(1)(A) and 1201(b)(1) with respect to the dynamic non-literal elements. The Ninth Circuit agreed with the lower court on the 1201(a)(1)(A) claim, limiting MDY’s liability for Glider to the dynamic non-literal elements.

According to the Ninth Circuit, because MDY “traffics in a technology … that is primarily designed, produced, or marketed for, or has limited commercially significant use other than circumventing a technological measure that effectively controls access to a copyrighted work,” it is liable under §1201(a)(2).

It would be interesting to explore further the question of whether these dynamic non-literal elements are sufficiently fixed in a tangible medium of expression to warrant copyright protection in the first place. I leave that exercise to the enterprising law student (hint: there is always the RAM copy). Ultimately, we may not know enough from the opinion about the technological underpinnings to make a determination one way or the other.

1201(b)(1). As for the §1201(b)(1) claim, the Ninth Circuit overturned the district court and held MDY (and Donnelly) not liable. Glider users are already authorized by the Blizzard agreements to copy the client software into RAM and such users do not exercise any other rights under copyright with respect to Blizzard’s works. Recall that §1201(b)(1) requires circumvention of a technological measure that protects a copyright owner’s right.

The Ninth Circuit blazed a somewhat new trail here, finding that §1201 creates a “new form of protection” separate and apart from copyright protection. Other federal courts—most notably the Federal Circuit—have declined to go that far, linking a 1201 cause of action to copyright infringement. Although this somewhat lengthy discussion in the Blizzard opinion is interesting for copyright jurisprudence, it is beyond the scope of this blog post. Refer to the opinion for more detail.

Conclusion

Sometimes the law should step in and regulate disputes between parties with adverse interests. But sometimes policy dictates that the parties work it out for themselves. In this case, the Ninth Circuit took more of a hands-off approach than the lower court.

Although the Ninth Circuit deemed MDY’s distribution of the Glider software a violation of DMCA §1201(a)(2), leaving MDY and potentially Donnelly with substantial liability, the decision did help clarify some questions of law and further deepened the circuit split, making a grant of cert by the Supreme Court all the more likely.

Website Immunity Under the CDA §230 in the 9th Circuit: Veering Toward Uncertainty

Prior to 2008, there was little doubt about the protection offered to websites by Section 230 of the Communications Decency Act of 1996 (CDA). In general, if a website (an “interactive computer service”) merely hosted or republished third-party content then the owners and operators of that site could shield themselves from liability for claims of defamation, fraud and other non-intellectual-property causes of action in which publishing is an element. For example, §230 generally protects Google against claims that advertisements appearing in its AdWords service are fraudulent (although that outcome is less certain in light of the Zynga case—see below).

Then, in 2008, the Ninth Circuit issued its decision in Fair Housing Council of San Fernando Valley v. Roommates.com in which it blurred the lines distinguishing “interactive computer service” from “information content provider.” In that case, because Roommates.com solicited potentially discriminatory answers in its roommate locator service, it could be held liable as an information content provider for discrimination under the Fair Housing Act and certain California housing discrimination laws as a result of discriminatory statements posted by users. In the opinion of the Court in Roommates.com, if there was illegal activity arising from user responses to Roommates.com questions related to race or gender, Roommates.com materially contributed to any subsequent discrimination. Many have questioned the reasoning of the majority in that en banc decision.

Although immediately subsequent decisions from courts in the 9th Circuit relied on Roommates.com to shield websites from liability for third-party content, more recently such courts have begun to characterize certain actions on the part of online services as those of a content provider.

Two recent cases illustrate this trend.

Zynga

In Swift v. Zynga Game Network, Inc., the District Court for the Northern District of California denied defendants’ (Zynga and Adknowledge) motion to dismiss claims of, inter alia, violation of California’s Unfair Competition Law (UCL) and the Consumers Legal Remedies Act (CLRA). Adknowledge provided ads to Zynga, who then displayed those ads to users of its popular online games, including FarmVille, Mafia Wars, and YoVille. Plaintiffs alleged that many of those ads were scams (for example, requiring you to provide your cell phone number and then billing you unauthorized charges via your cell phone bill).

The defendants moved to dismiss the CLRA and UCL claims under §230 because they were merely republishing the allegedly illegal ads provided by a third-party. Citing the Ninth Circuit’s holding regarding material contributions by service providers in Roommates.com, the Court here denied defendant’s motion, in part because the ads were tied to virtual currency that could be used in Zynga’s games, and because Zynga was responsible for “the design, layout, and format of the special offers.” Furthermore, the Court reasoned, these special offers “appear directly within Zynga’s games.”

Even under the most generous reading of Roommates.com, it is not clear how the virtual currency or the design and layout should affect the application of §230. Unlike the content-specific questions posed on the Roommates.com site, the virtual currency has no logical connection to the special offers, except that it might create a stronger incentive for Zynga users than the average internet ad. As for the design and layout, this is a dangerous peg upon which to hang liability since every website has implemented some sort of design and layout. If that were a crucial factor, more websites containing third-party content would be denied the benefit of CDA immunity.

This case appears to conflict with the holding in Goddard v. Google, a decision out of the same court from 2009 (mentioned briefly above). The Zynga court attempted to distinguish the Goddard case by stating that Google merely suggested keywords to fraudulent advertisers. However, I fail to see how this is further removed from content creation than Zynga’s virtual currency.

Kruska

In Kruska v. Perverted Justice Foundation Incorporated.org, the District Court for the District of Arizona denied defendant Brocious’ motion to dismiss a claim of defamation under §230 because, as far as we can tell from the opinion, plaintiff accused defendant in her pleading of “actively contribut[ing] to [the websites’] content.” That is the only reasoning we can glean from the published opinion, making it a somewhat frustrating read; it is so thin on relevant facts, and in its reasoning. Without digging deeper into the background of the case, the Court leaves us with almost no guidance as to what this means for §230 jurisprudence.

Granted, these are just denials of motions to dismiss under 12(b)(6). Defendants may still prevail at the summary judgment stage. But part of Congress’ intent in passing §230 was to save defendants like Brocious, Zynga and Adknowledge from the costs and hassle of litigation where they played no role in the creation of the offending content.

It is unfortunate that these cases introduce uncertainty into a statute that could benefit from more judicial clarity. Without a clear line as to what constitutes the creation of content, website owners will have a more difficult time determining whether the design of a website or their activities in relation to user content will lead to liability.

As we can see from the conflicting Zynga and Goddard cases, post-Roommates.com courts are struggling to figure out when a service provider is materially contributing to the creation of content. Maybe it’s time for the Ninth Circuit to hear an appeal and clarify the test.

Potential Conflicts Between the Apple App Store Terms and the GPL

Every so often someone asks me whether it is possible or “legal” to use GPL code inside an iOS/iPhone/iPad app, whether the iTunes App Store terms and conditions permit open-source licensing within the app. My understanding is that Apple won’t block an app that contains open-source code but will remove an app if the copyright owner of the open-source code complains that the terms of the applicable open-source license are being violated, typically due to a conflict with App Store terms. For this reason it is important to understand whether a particular license would be in conflict with the App Store terms.

Recently, Brett Smith, a compliance engineer with the Free Software Foundation (FSF), posted to the VLC media player mailing list his thoughts on the compatibility of the GPLv2 with the current iTunes terms and conditions.

First of all, Apple’s terms do allow for the application of a developer’s own license agreement to the app:

Your license to each App Store Product is subject to the Licensed Application End User License Agreement set forth below. You agree that the terms of the Licensed Application End User License Agreement will apply to each Apple Product and to each Third-Party Product that you license through the App Store Service, unless the App Store Product is covered by a valid end user license agreement entered into between you and the licensor of the App Store Product (the “Licensor”), in which case the Licensor’s end user license agreement will apply to that App Store Product.

So, even though the FSF does not consider the GPL to be an “end user license agreement,” Smith concludes that Apple likely didn’t have that distinction in mind and was likely not trying to exclude licenses such as the GPL. Therefore, the App Store terms do leave room for open-source licenses in this section of the agreement.

However, the section in the App Store terms titled, “USE OF PRODUCTS AND THE SERVICES” states that end users must use Products (including apps) in accordance with certain Usage Rules. Altogether there are nine such rules, only two of which apply to apps:

(i) You shall be authorized to use Products only for personal, noncommercial use.

(ii) You shall be authorized to use Products on five Apple-authorized devices at any time …

Because Section 6 of the GPLv2 requires that licensees place no restrictions on use or distribution, these Usage Rules likely conflict with the terms of the GPLv2.

That answers the question regarding code licensed pursuant to the GPLv2, at least from a conservative, FSF perspective. But what about version 3 of the GPL? Similar to GPLv2, GPLv3 prohibits “further restrictions” (§7). Given the language in that section of the GPLv3, Apple’s Usage Rules would likely be deemed further restrictions and therefore also in conflict with the terms of the GPLv3.

And, keep in mind, both versions of the GPL impose obligations on the licensee (e.g., making source code available, including certain attribution notices, providing a copy of the license) with which the iOS app developer must comply. Compliance with those obligations is beyond the scope of this post.

What about other open-source licenses? Although most “academic” open-source licenses would likely not conflict with the App Store terms, whether the App Store terms conflict with other copyleft licenses (e.g., the MPL or the CPL) depends on the terms of those licenses.

Again, in practice, based on what we know about the app approval process, whether use of open-source software in an iOS app will be permitted depends in large part on whether the copyright owners of the open-source code (a) are aware of the use and (b) file a complaint with Apple. Regardless, if an iOS app developer wants to be a good open-source citizen, he or she should consider potential conflicts before deciding to license code as open-source or use code made available under an open-source license.

Will Apple eventually accommodate open-source software more explicitly in its apps ecosystem? It seems like the biggest impediment for Apple would not be the licensing terms but its software DRM technology. Making special exceptions for open-source apps could be more trouble than it’s worth for Apple.

What Does “Open” Mean Exactly?

It’s cool to be open these days. Companies tout their openness and the openness of their technologies. But not everyone understands, or can agree on, what “open” means, in part because companies like Google would prefer to define it how they see it, to make it synonymous with all that is good and right in the world, regardless of historical meaning.

For example, on a lot of tech podcasts and a few blogs lately, I’ve heard technology pundits conflate open source with a different meaning for “open,” which is that used in the phrase “open standard” or “open platform.” They are certainly different concepts, even though there is some overlap at times. Leo Laporte is just one pundit guilty of conflating these concepts.

In the wake of attempts by Apple and Google to redefine the word for marketing and PR purposes it’s important to remember the distinction. So I will add my $0.02 to the conversation and explain how the concepts around “open” differ.

In order to reign in the scope of this blog post, I’ll focus primarily on the recent kerfuffle involving Android versus iOS as a lens through which to view the different meanings of “open.” I need to cover a lot of ground so please bear with me.

Apple Earnings Call

On a recent Apple earnings call, Steve Jobs talked about the downside to the openness of the Android platform, at one point citing Windows as an example of a more “open” platform. Many took issue with this because iOS is a closed platform and accused Steve of using Orwellian “double-speak.”

Here is Jobs:

In reality, we think the open versus closed argument is just a smokescreen to try to hide the real issue, which is, “What’s best for the customer—fragmented versus integrated?” We think Android is very, very fragmented and getting more fragmented by the day.

Not surprisingly, Jobs continues to believe very strongly in the vertically-integrated, closed approach, saying, “We think this is a huge strength of our approach compared to Google’s. When selling to users who want their devices to just work, we believe integrated will trump fragmented every time.” (emphasis added) To Jobs, open equals fragmented. “It just works” has been Apple’s mantra since its resurgence in the late 1990s. However, Jobs takes issue with the word “closed” being applied to what he thinks of as “integrated.” Perhaps it’s just a matter of nomenclature.

Again, Jobs:

Google loves to characterize Android as “open,” and iOS and iPhone as “closed.” […] The first thing most of us think about when we hear the word “open” is Windows, which is available on a variety of devices. Unlike Windows, however, where most PCs have the same user interface and run the same apps, Android is very fragmented. Many Android OEMs, including the two largest, HTC and Motorola, install proprietary user interfaces to differentiate themselves from the commodity Android experience. The user’s left to figure it all out. Compare this with iPhone, where every handset works the same.

Nevertheless, as I define the meaning of “open” in this post, we will see that things are not as simple as either Google or Apple suggests. iOS is a mostly closed-source, partially-closed platform that supports some open standards. So is Android.

Android

Android is Google’s operating system for mobile devices that’s based on the Linux operating system and developed by Google and the Open Handset Alliance. There are over 20 smartphones available in the US running Android, and another 90 or so worldwide. Google claims that Android is open. From what I can tell, Google means “open” both in the open source sense of the word and the open platform sense.

Apple and Google have strikingly different approaches to the mobile phone market. It is reminiscent of the MS Windows versus Apple days of yore, with Google’s Android platform standing in for Windows. After all, there is only one iOS smartphone currently available: the iPhone 4. Of course Apple, Inc. is not the Apple Computer of the 80s and 90s, although it has stood firmly behind its belief in a vertically integrated approach to computing. And Google is hardly the Microsoft of smartphones. Most agree that, unlike with the PC, there will be no dominant company or platform for mobile devices.

Let’s explore the different types of openness, starting with open standards.

Standards

What is a standard? In the technology context, it is a description of engineering processes or technical criteria that allows different parties to develop technology so that it works predictably with other technology. Standards are sometimes developed by a group of interested parties—companies and individuals collaborating via a standards organization—and published so that they can be of use. These are known as open standards (see below). Other times a standard arises organically through market forces; these are de facto standards.

The benefits of standards are summarized well by Treacy and Lawrance:

Industry standards are widely regarded as beneficial. Subject to appropriate rules, standards promote innovation by circumventing problems of product compatibility (particularly important for technologies reliant on interfaces and networks), they disseminate chosen technologies by allowing access to all comers, and they enhance technology durability.

De Facto Standards

A de facto standard is a standard or interface that was not developed through a standards-making process but rather through the sheer popularity of a single product or line of products. Examples include everything from the QWERTY keyboard to the MP3 music format to the PDF (first made popular by Adobe). As we see in the case of PDF, de facto standards sometimes become formalized standards.

Examples of Standards

Standards permeate our daily lives in the modern world, and have since the industrial revolution: the 1/2-inch (12.7mm) spacing of the rollers on a bicycle chain (started out as a de facto standard); alternating and direct current electricity; lamp bulb socket sizes; and VHS tapes (winning out over Betamax). More recently, HTTP and SMTP are open Internet standards that enable websites and email respectively. Likewise, AAC (digital compressed audio) and the DVD spec are standards, although less open.

Open Standards

In contrast with a closed or de facto standard, open standards “allow market participants to share knowledge and develop ‘best-of-breed’ products.” (citing a 2006 DOJ Business Review Letter) Determining whether a particular standard is open is not a simple matter. It all depends on who you talk to. There are many definitions. However, the essence of these varying definition is this: an open standard is a publicly available standard that creates a fair, competitive market for implementations of the standard. Often, an open standard may also have been developed through an open process.

The Wikipedia entry for “Open Standard” contains a very comprehensive list of different definitions for the term, from a fairly diverse cross section of the relevant IP and tech policy experts and organizations. Reading that list you’ll notice that there is a whole spectrum of openness when it comes to technical standards.

Some of the most common features of an open standard are as follows:

• Developed, approved and maintained through a collaborative and consensus-driven process, which is reasonably open to all interested parties.
• Intellectual property (IP) rights necessary to implement the standard are licensed by the rights holders on a fair, reasonable and non-discriminatory (FRAND) basis.
• No discrimination as to implementor or implementation, although the standard can require certification (typically so long as it is low- or zero-cost).
• By creating a fair, competitive market for implementations of the standard, they do not lock consumers into just one vendor or group.
• Sometimes open standards prohibit predatory practices, such as embrace-and-extend. “Embrace and extend” (sometimes “embrace, extend and extinguish”) is a term that was coined within Microsoft some time in the 1990s to describe what was a fairly common Microsoft practice of publicly embracing open or widely-used standards and then extending those standards with proprietary features in order to lock in users. Microsoft actually has its own definition, which is roughly in line with the rest.

Regarding the effect of the FRAND obligation, much has been written. This subject warrants an entire blog post in itself so I will only mention a few common interpretations here. FRAND often requires, among other things:

• parties submitting technologies to be included to state a maximum royalty for any relevant patents;
• sanctions for failing to disclose patents necessary for the standard; and
• limits on contract terms such as grant-backs, non-asserts and covenants not to sue (anything that would be discriminatory in other words); and
• creation of a public licensing policy.

Although some companies may balk at being subject to FRAND obligations, there are potentially numerous benefits from doing so, including lower-friction negotiations, reciprocal treatment, and decreased risk of an antitrust or contract breach lawsuit.

The Open Source Initiative (OSI), the non-profit that oversees the Open Source Definition and blesses licenses as open-source, only deems a standard open if it does not preclude open-source implementations. In addition, with respect to open-source software, many open-source licenses contain a “grant-forward” patent license provision or a patent defense clause in an attempt to render the open-source software immune to patent lawsuits. As of this writing, the validity of these types of clauses in open-source licenses is undetermined.

As you can see, the open source concept and the open standard concept sometimes intersect; an open standard may require an open-source implementation.

Standards Setting Organizations

A standard setting organization (SSO) provides a forum where companies collaborate in the drafting of new (typically open) standard. Crucial to the focus of this post, many SSOs impose certain rules and procedures on the participants regarding intellectual property rights. For example, participants may be required to license relevant patents under FRAND (as we saw in some of the example definitions above).

It is important to note that SSOs are voluntary associations, typically having no power to enforce the rules they promulgate. Nevertheless, they are taken seriously, in part because development of a standard via committee has been proven to be more efficient than through purely private interests operating in the market. Open standards create more value overall than closed standards, making it a win-win for participants in the long-term. Technology is complicated and innovation often requires cooperation. So an open standard is more likely to survive than a closed standard. In addition, because competition and antitrust laws often apply to SSO participants, there are real costs to behaving unethically as an interested party.

Well-known examples of SSOs include the Institute for Electrical and Electronic Engineering (IEEE), which facilitated the creation of the 802.11 WiFi standard and the Unix POSIX standard, and the Internet Engineering Task Force (IETF), which facilitated the promulgation of many of the Internet protocols on which we rely today.

Why Go Open?

Companies may choose to participate in an open standard and offer their expertise, research, and intellectual property to other participants for a number of reasons. Open standards usually produce more value than closed standards and can better guarantee the survival of a technology in what would otherwise be an intensely competitive device or technology market.

Although being part of an open standard potentially increases coordination on a new technology (thereby increasing its total value), as Tim Simcoe points out, it can also “increase the intensity of competition, making it harder to capture that value once the new standard is introduced.” Consequently, companies may be tempted to game the standards setting process or encumber it in order to obtain at least a short-term advantage. The only alternative for those who see the long-term benefits and participate fully in a standard is competing on implementation, a well-known phrase in the industry which means that participants in open standards can extract value from the standard by using “time-to-market advantages, secrecy, superior design and marketing, or production cost advantages.” Apple is just one example among many that competes using a hybrid of closed platforms and open standards by competing on implementation.

As you can see, what constitutes an open standard is subject to interpretation and far from clear-cut. Standards setting is often contentious and wrought with uncertainty. In the United States, courts have only begun to offer guidance regarding appropriate behavior for patent holders implicated by open standards. In the US, such lawsuits are initiated by the Federal Trade Commission (FTC) under antitrust law. Unsurprisingly, some of the litigated issues include (a) whether a patent declared essential to a standard deprives the patentee of a right to an injunction where it does not license the patent on fair, reasonable and non-discriminatory (FRAND) terms; (b) the meaning of the word ‘essential’ and (c) the meaning of FRAND.

Open Platforms

Although sometimes related, open platforms do not necessarily contain or rely upon open standards (nor open-source software). An open platform is a software system (e.g., an operating system) whose application programming interfaces (APIs) are published and available to developers on the system or platform.

Given this definition, Linux, Mac OS X (so far), Android and Microsoft Windows are all good examples of open platforms. And Linux is a great example of how an open platform can also be open-source and an open standard (it is POSIX-compliant). Although open platforms do implement many open standards (e.g., 802.11 networking, TCP/IP), they are typically not designed through an open-standards process. POSIX-compatible flavors of Unix (such as Linux) are one exception, although the standardization of the platform came long after its initial design.

So, although Apple acts as the gatekeeper for apps on the iOS platform, under the traditional definition of the phrase, iOS is an open platform; its APIs are published to developers. However, some would argue that it is a closed platform since users are not free to install software of their choosing. And it is this narrower meaning of “open platform” that Google is trying to leverage to its advantage, in portraying Android as more “open” than iOS.

Open Source

In contrast with open standards and open platforms, open-source software (OSS) is software that is licensed under a special kind of software license agreement, one that typically permits (royalty) free redistribution and modification, offering access to the source code for the software and sometimes requiring that licensees make modifications available under the same license terms. The Open Source Initiative (OSI) offers a more comprehensive definition.

Some open-source software is also known as “free software,” which is a narrower category of OSS in which the license requires any modifications to be licensed under the free-software license. The GNU General Public License (GPL) is one example of a free-software license.

The most common examples of OSS are the Linux operating system, the Firefox browser, and the Apache HTTP server. Each of these are examples of software that implement one or more open standards. In fact their existence relies heavily on the adoption of the open standards that they support.

OSS in iOS and Android

This brings us back to iOS and Android. Some might say that iOS is as open as Android, once you take into account the relative ratio of closed-source, proprietary software to open on each platform. After all, iOS is based on Mac OS X, which contains the open-source Darwin kernel, as well as many other open-source remnants of BSD code. Darwin is licensed under the Apple Public Source License (APSL), a copyleft license similar to the Mozilla Public License (MPL). Furthermore, the Safari rendering engine (WebKit) is open-source, as is Apple’s Bonjour networking protocol software, both of which ship on iOS devices.

Regarding Android, the Android software stack is licensed pursuant to the Apache License, 2.0 (for the most part). The open-source parts of Android are licensed pursuant to either the Apache License 2.0 or the GNU General Public License, version 2 (GPLv2).

However, many marquee features of the Android platform are made available via proprietary (not open-source) software owned by Google, including: the Android Market (app store), IM client, gmail client, Google Maps, YouTube app, calendar app, setup wizard, and contacts syncing framework.

Furthermore, Google does not make the most recent version of even the open-source components in Android available to the public until some time after the new version has gone out to handset manufacturers. Some developers find that this discourages community contributions to the open-source Android project.

What this all comes down to is whether Android is a more open platform than iOS.

Is Android Open?

So, what about Android’s claim to openness? Does it matter that it’s an open platform in the narrower sense that Google prefers? Many techies and software developers believe that it does matter, that this is what makes Android superior and destined to dominate over iOS. After all, anyone can create an Android app and any user can install that app. But, this is also a weakness and may not be entirely accurate.

MG Siegler, an admitted “Apple fanboy,” emphasizes the downside of an open platform: that it can be co-opted by carriers and handset manufacturers, resulting in a negative user experience. The ultimate irony is that many carriers and handset manufacturers do make Android more closed, as AT&T did with the Motorola Backflip and the HTC Aria, whose feature sets are locked down. And Verizon recently locked the Samsung Fascinate so you could only use Bing as the search engine. All of this makes one wonder what it would take to preserve the openness of a mobile platforms when so many hardware manufacturers and carriers have interests that may compete with an optimal, truly open user experience.

Most troubling to the gotta-have-it early adopters out there is that carriers usually delay rollout of Android updates significantly. Although the exact figures are disputed, it appears that roughly 70 percent of Android phones still aren’t running the latest Android OS 2.2 (Froyo) and almost 30 percent aren’t even on version 2.x.

Android’s reputation as being “open” can lead to developer confusion as well. When a developer recently modified the Android image and made it available to handset owners, Google served him with a cease-and-desist letter to stop his redistribution of the proprietary apps (Google Maps, gmail, etc.).

Furthermore, the Android handset market does appear to be quite fractured. According to Steve Jobs, there are “a hundred different versions of Android software on 244 different handsets.” This introduces a challenge for some developers. Recently, Greg Peters from Netflix posted explained why Netflix has yet to release the Netflix app on Android (note that it is already available on Windows Phone 7). As Peters explains (emphasis added):

The hurdle has been the lack of a generic and complete platform security and content protection mechanism available for Android. The same security issues that have led to piracy concerns on the Android platform have made it difficult for us to secure a common Digital Rights Management (DRM) system on these devices… . Although we don’t have a common platform security mechanism and DRM, we are able to work with individual handset manufacturers to add content protection to their devices. Unfortunately, this is a much slower approach and leads to a fragmented experience on Android, in which some handsets will have access to Netflix and others won’t.

According to Peters, since “providing the [Netflix] service for some Android device owners is better than denying it to everyone,” Netflix will release the Netflix streaming app on “select Android devices” in early 2011.

There is no doubt that different Android users are having different product experiences. Sometimes a “sea of choice” is not a good thing for consumers already overwhelmed by choice in the smartphone market. Or is it? Android adoption only continues to accelerate, rising to second among smartphone operating systems in the most recent quarter.

Is the Device the Appropriate Focus?

Mike Elgan over at Datamation makes a different point about Android’s relative openness. According to Elgan, it is essential to focus on Google’s core business in assessing the openness of Android. Google’s business is still primarily selling ads; no one would dispute that. Approximately 99 percent of Google’s revenue comes from advertising. Google is not a consumer products company. By contrast, Apple is in the business of selling hardware and this is where they make money. In both cases, the companies are “very closed, secretive, and controlling about the part of their business that makes the money.” This is an interesting point but perhaps too much of an over-simplification in its analogy. Google is making the openness argument with the apps ecosystem clearly in mind, to highlight Apple’s overly maternalistic and tight control over its iOS platform. Nobody (that I am aware of) is clamoring for more open access to Google’s ad networks. Moreover, Google’s ad services are exactly that, hosted services. Android and iOS are distributed software that actually can be extended by end users to some extent. But it does further highlight the framing that Google is employing here.

Conclusion

Chris Grams wrote an insightful post over at opensource.com about how defining “open” and draping yourself in the open flag (my words not his) is a way to position or reposition your company more favorably in the market, especially a market like smart phones, where techies are the early adopters that set the pace of adoption for a new product. As we saw with some of the tech pundits cited above, mere invocation of the word “open” gains a market laggard huge points in the race for mindshare. This appears to be working for Google to some extent.

The battle of business models—open versus closed—is a fascinating battle to watch. Many are predicting that the iPhone and iPad will go the way of the 1980s Apple, i.e. with the more “open” platform sidelining Apple. To understand the battle though, it helps immensely to understand who is really open and exactly where the openness matters, where it is the true reason for an advantage.

The smartphone and mobile computing market has the most potential for growth; it’s where all the important technology battles are being fought. In fact, some predict that the mobile computing market will be “larger than the PC market ever was.” So the nomenclature matters. It is a battle of words and ideology. Steve Jobs would like us to believe that partially closed is preferable to fragmented, and leads to a superior user experience. Eric Schmidt would like us to believe that choice and developer freedom is what consumers want.

Open standards are clearly a benefit to consumers. In a similar fashion, there are instances where open-source software is a benefit to consumers, although mostly on the back-end. When it comes to end-user software, Firefox is the rare example of a quality user experience from open source. But are open platforms inherently better?

I don’t know, maybe these are distinctions that only matter or make sense to lawyers and other similarly afflicted individuals. To techies it appears that open is open is open. Not only that but the whole open distinction is lost entirely on the average consumer. And most of these phones, Android or not, are sold to the average consumer, not some geek who wants to mod the GUI and gain root access. So the question is whether this “openness” has secondary effects like an overall better consumer experience, better apps, and lower overall costs. As of this writing, there is no clear winner. Maybe there never will be.

Update 2010-11-26: Tim Bray has put together a nice little graphical tour through the Android architecture.

Client Lists Likely Not Trade Secrets in an Era of Search and Social Networks

I don’t talk about trade secret law much on this blog. It is the kid on the intellectual property playground who never gets picked for touch football. In general, there are far fewer cases compared with patents, copyrights, and trademarks. And the fact patterns are often far less sexy, dealing with mundane business information such as contact lists and marketing plans. But Sasqua v. Courtney does contain some interesting take-aways.

Can a client contact database that is easily replicated by performing Internet searches be protected as a trade secret? The obvious answer is no. But it took the United States District Court for the Eastern District of New York forty-one pages to come to the same conclusion. The ultimate holding in this case is relatively fact-specific, making its applicability to other misappropriation scenarios uncertain. However, as a fairly thorough walk-through of the relevant factors in determining whether information is a trade secret under New York law, the opinion serves as a good reminder of some best practices.

The case is nevertheless important because it illustrates the effects that Internet search tools and social networking have had on the protectability of certain types of information as trade secrets and underscores the importance of safeguarding valuable information properly, according to the applicable state law standard in your jurisdiction.

Facts

In Sasqua Group, Inc. v. Courtney, the owner of Sasqua, an executive recruiting agency specializing in the placement of professionals in the financial services industry, sued Lori Courtney, a former independent contractor of the agency, for trade secret misappropriation for Courtney’s alleged use of a Sasqua client contact database. Courtney’s new venture is Artemis Consulting, a competing recruiting firm that Courtney founded in late 2009 or early 2010. Prior to starting Artemis, Courtney worked with Sasqua for about ten years, eventually earning the title “Managing Director.” The founder and owner of Sasqua is a man named Christopher Tors.

In support of its recruiting and placement business, Sasqua maintained a central database of client information, including “client contact information, individual profiles, contact hiring preferences, employment backgrounds, descriptions of previous interactions with clients, resumes and other information.” This database was populated by Tors, Courtney and others and maintained by a computer technician named Douglas Steinschneider.

When Courtney left Sasqua earlier this year to start Artemis, Tors assumed she was using the Sasqua client database, in part because several soon-to-formed Sasqua clients had already been contacted by Courtney when Tors contacted them to share the news of Courtney’s departure.

In the complaint, plaintiffs claimed trade secret protection for “the confidential proprietary and competitively sensitive information about Sasqua’s client contacts, their individual profiles, their hiring preferences, their employment backgrounds, and descriptions of previous interactions with client contacts.” According to the plaintiffs, such information was not publicly available and could not easily be duplicated.

Courtney’s position regarding Sasqua’s client list was that Sasqua employed no “particular or proprietary methodology” to perform its searches. To the contrary, anyone could create a comparable contact list in the capital markets industry by performing searches at Bloomberg, LinkedIn, Facebook and other “publicly available databases.” Moreover, argued Courtney, “the contact information that a search firm may assemble in a database is almost immediately obsolete.”

Trade Secrets Generally

A trade secret is, in essence, information not generally or publicly known for which reasonable precautions have been taken to protect it from public disclosure. If someone uses improper or wrongful means to discover a trade secret, that person can be liable for trade secret misappropriation. Improper or wrongful conduct is that which falls below generally accepted standards of commercial morality and reasonable conduct. Reverse engineering is not typically considered improper in the absence of a binding non-disclosure agreement (NDA).

Unlike patent, trademark and copyright law, trade secret law is a matter of state law. As a result, it can vary slightly from state to state. Like some areas of contract law, there have been attempts to harmonize the various state laws, which is why we have the Restatement (Third) of Unfair Competition as well as the more widely accepted Uniform Trade Secrets Act (UTSA). The UTSA has been adopted by a majority of states (e.g., California Civ. Code §§3426 et seq.), although New York is not among those states. Nevertheless, the principles are generally the same in UTSA and common law trade secret law states.

The UTSA defines a trade secret as information that:
1. derives independent economic value, actual or potential, from not being generally known to the public or to other persons who can obtain economic value from its disclosure or use; and
2. is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.

New York Trade Secret Law

Under New York law, which is modeled on Section 747 of the Restatement of Torts, a trade secret is “any formula, pattern, device or compilation of information which is one uses in his business and which gives him an opportunity to obtain an advantage over competitors who do not know or use it.” New York courts apply several factors in determining whether information constitutes a trade secret, including:
1. the extent to which the information is known outside of the business;
2. the extent to which it is known by employees and others involved in the business;
3. the extent of measures taken by the business to guard the secrecy of the information;
4. the value of the information to the business and its competitors;
5. the amount of effort or money expended by the business in developing the information; and
6. the ease or difficulty with which the information could be properly acquired or duplicated by others.

Because these factors were first enumerated in this form in Ashland Management v. Janien (1993), they are known in New York as the Ashland Management factors.

In New York, a plaintiff must prove that: “(1) it possessed a trade secret and (2) defendant is using that trade secret in breach of an agreement, confidence or duty, or as a result of improper means.” In addition, for the purpose of deciding whether to issue a preliminary injunction, a New York court may presume irreparable harm where a trade secret has been misappropriated.

There have been a number of cases in New York finding a protectable trade secret in customer lists developed through “substantial effort and kept in confidence.”

The Ashland Management Factors

In order to determine whether the client list at issue in this case was a trade secret, the United States District Court for the Eastern District of New York considered each of the Ashland Management factors in turn.

1. Is the Information Known Outside the Business?

Courtney argued that, given the ease of recreating the information in the Sasqua list using Internet search tools, the information is effectively known outside the business. In fact, it’s not entirely clear from the opinion that Courtney copied the Sasqua database.

The Court recognized that there may be industries or recruiting searches in which assembling a contact list is like looking for a needle in a haystack—e.g., where there are only a handful of engineers in the country with a certain expertise—but stated that this is clearly not the case for Sasqua’s clients, partly due to Courtney’s impressive ability to quickly recreate the client list from publicly available sources. Potential clients in the financial services recruiting field are easy to identify.

Courtney’s Process

In order to recreate the client database anew at Artemis, Courtney started by performing an Internet search, searching for “high turnover at banks.” Once she had that information, she would perform a search on Bloomberg, LinkedIn, or Google. Courtney sometimes consulted FX Week, a publication containing “substantial information about foreign exchange businesses and … emerging markets.” Since FX Week has a searchable database, Courtney could further narrow her searches regarding decision makers for financial institutions. Courtney was often able to locate a workable phone number using Google, Bloomberg or LinkedIn. With this information regarding a bank or business in hand, Courtney would then call whomever she deemed a potential decision-maker, attempting to establish her expertise in the field and a rapport with the contact. According to Courtney, this whole process typically took no more than five minutes. It was that simple.

Of course plaintiffs argued that the process of locating the right decision-maker was not nearly so simple. But plaintiffs failed to produce any evidence to that effect. Plaintiffs also asserted that its misappropriated trade secrets were the combination of client contacts and those contacts’ preferences and placement histories. However, the Court reasoned that Courtney could simply ask potential or existing clients about their preferences.

During Courtney’s testimony in the case, she described a changing recruiting landscape, a “post-Lehman bankruptcy world” in which candidates were far less protective of their personal information and, due to panic and a desire to be connected, voluntarily publish their contact information online. Reading the opinion, one sees that the Court was heavily swayed by this characterization in finding that a simple recruiting contact list is not protectable as a trade secret.

2. Extent of Employee’s Knowledge

None of the recruiters working for Sasqua signed confidentiality agreements. And the client database was freely and widely disseminated to all those associated with Sasqua. The Court therefore decided this factor in favor of Courtney and against a finding of trade secret protection.

3. Measures to Protect Secrecy of the Information

According to the Court, this is the most important of all the Ashland Management factors. After all, it does comprise half of the definition of “trade secret” in the UTSA: the owner of a trade secret must “take reasonable measures to protect its secrecy.” The Court cites some of the measures approved by courts: “use of passwords, alarm systems, and installation of firewalls and security software.” In this case, Steinschneider testified that Tors asked him to leave the database on Courtney’s computer after she gave Tors notice of termination, the database was not password protected, the database was freely distributed to temporary workers, many former Sasqua recruiters retained copies of the database with Sasqua’s knowledge, and the database did not contain any “legends designating confidential and proprietary information” as a reminder to recruiters. Steinschneider further testified that the Sasqua database was initially misappropriated by Tors from his prior employer, causing the Court to consider an unclean hands argument in the alternative.

Courtney testified that many who never worked for Sasqua have copies of the Sasqua database, including a group in London with whom Sasqua had considered a joint venture.

In short, the Court reasoned, “Sasqua failed to take even basic steps to protect the secrecy of the information contained in the database.”

4. Value of the Information to the Business and Its Competitors

The Court focuses primarily here on two things: (1) the fact that the Sasqua client database was based in part on a client database that Tors took from a prior employer and (2) the fact that, according to both Courtney and Steinschneider, the Sasqua database was “exceedingly old and stale, much of it … never accessed let alone used.” Apparently, those who make up the financial industry workforce change employers often, making such a database inherently difficult to maintain. In the words of the Court, “whatever value to Sasqua the database information once possessed clearly diminished over time as its utility to Sasqua’s consultants declined.”

5. Time and Effort Expended in Developing the Information

Unlike copyright law, where mere “sweat of the brow” (hard work in assembling information) is not protectable as such, trade secret law considers the effort expended by a plaintiff in determining whether information is a trade secret. For example, if an organization must develop a reputation with military officials and families over a long period of time, this may support a finding of trade secret protection for a database related to such relationships. However, in this case, the client database was not “gained through extraordinary effort or expensive research.” Portions of it were copied from a prior employer, much of it was out-of-date. And Courtney demonstrated how little effort was required by showing how easy it was to recreate the database from scratch (see above).

6. Ease With Which the Information Could Be Properly Acquired or Duplicated

As we have already seen, the information utilized by Courtney in competing with Sasqua was easily acquired. Most of it is publicly available, at least to those with the requisite skills.

Preliminary Injunction

In deciding whether to issue a preliminary injunction, the Court deemed it especially relevant that Courtney was merely using the purportedly secret client information and not disseminating it to third-parties. In New York, there is precedent stating that a preliminary injunction is not necessary where there is no further dissemination. Part of the reasoning here is that, if the information really is valuable due to its secrecy, the defendant has as much incentive to maintain the information as confidential as the plaintiff.

Lessons re Confidentiality

Despite its fact-specific nature, there are several aspects to this case that are of general applicability to any company attempting to protect its confidential information as a trade secret. They are as follows:

• If you are going to expend any time, effort or expense in protecting information, make sure it is information that is not publicly available using Internet search tools or other means.
• If the information you seek to protect is a client list, be sure it is something that requires substantial investment to create and maintain, that requires some specialized knowledge or expertise to assemble.
• Once you have created such a list, take appropriate measures to maintain its secrecy:
  • Encrypt and password protect the database.
  • Limit who you provide the database to.
  • Maintain the database behind a network firewall.
  • Have your IT department implement secure laptop computer policies that include the ability to remotely wipe a lost or stolen computer.
  • Require those who have access to enter into a binding agreement containing confidentiality terms.
• Be sure the information is updated such that it remains relevant and current.

Finally, be sure to add value and compete in ways that ensure that the success of your company does not rest solely on the secrecy of a client list. Consider relying on non-compete and non-solicitation obligations in your employment and contractor agreements, to the extent such restrictions are enforceable in your jurisdiction.

Conclusion

Again, this case is highly fact-specific and it sets forth a trade secret test directly relevant only in the New York courts. Unlike some more technical industries, the executive recruiting business is “a comparatively low-technology business” where success is “driven by general managerial expertise as opposed to the application of highly technical, proprietary, or secret information.” But the case does underscore the importance of maintaining one’s valuable business information as a trade secret. And it highlights the changing nature of trade secrets as more and more information has become publicly searchable online. What may once have been a protectable trade secret no longer is.

As is the case with many industries, what used to be sufficient to compete may no longer be sufficient. What was once a business advantage may turn out to be something to which everyone has access. Like the music industry, recruiting firms must innovate to remain relevant in the networked world.

After reading this case, one is left with the impression that it was more about bad blood and bitterness on the part of the plaintiff than about any legitimate trade secret claim. The facts militate too far in Courtney’s favor for it not to be driven in large part by emotion.

Big Surprise: A Download Is Not a Performance Under Copyright Law

On September 28, 2010, the federal Court of Appeals for the Second Circuit confirmed that a download of a digital sound recording (e.g., an MP3 or iTunes purchase) was not a “public performance” under the Copyright Act. This is an important decision for online music sellers such as Apple and the defendants in this case: Yahoo! Inc. and RealNetworks, Inc. because they avoid paying license fees over and above what they are already paying for the right to offer digital music purchases/downloads. Although it may seem obvious to some readers that a download is not a public performance of the music contained in the download (in the same way that carrying a physical CD home from the record store is not equivalent to playing that CD in a club or on a radio station), ASCAP thought it was worth a shot, twice (at trial and on appeal). Let’s examine the arguments put forth by ASCAP and the Court’s sound reasoning. But first, a little background on the lawsuit and a very brief refresher on music copyright.

Recall that a song is composed of two separate copyrighted works: the musical composition and the sound recording. You can think of the composition as the sheet music (the song expressed therein anyway) and the sound recording as what was captured in the studio or on stage and printed on a CD or a digital file. Also recall that the licensee of a copyrighted work such as an MP3 need only license the copyright rights (copying, distribution, public performance) needed for that licensee’s particular use. For streaming, an internet radio station need only license the public performance right; for downloads, the rights to copy and distribute.

ASCAP is one of several performance rights organizations (PROs) that licenses the public performance right for musical compositions. Because ASCAP licenses roughly 45% of the musical works licensed for online streaming, a decision by the Court in favor of ASCAP would have made quite a splash for the Pandoras, last.fms, and Rdios of the ‘net.

Here, both Yahoo and RealNetworks were offering music via both online streaming and digital download. They were already paying ASCAP for the right to stream (i.e., publicly perform) the music in their catalogs. More specifically, both companies had “blanket licenses” from ASCAP, which gives them the right to perform all works in the ASCAP repertory for a single rate which does not vary with the amount of music actually exploited. But ASCAP wanted to extract additional license fees for the downloads, arguing that a digital download is a public performance. It is hard to understand ASCAP’s argument here; they rely primarily on a selective and implausible reading of the definition of “publicly” in §101 of the statute.

In reaching its holding that a download is not a public performance, the Court applied the first of the canons of statutory interpretation: focusing primarily on the plain meaning of “publicly perform” in §101, examining the definition of “perform” together with the definition of “publicly.” According to §101, to “perform” a copyrighted work means “to recite, render, play, dance, or act it, either directly or by means of any device or process.” Since, the Court reasoned, a download is neither a “dance” or an “act,” it must “fall within the meaning of the terms ‘recite,’ ‘render,’ or ‘play.’”

The Court then ran through each word in turn and provided examples of music being recited or played. My favorite:

Itzakh Perlman gives a “recital” of Beethoven’s Violin Concerto in D Major when he performs it aloud before an audience. Jimmy Hendrix memorably (or not, depending on one’s sensibility) offered a “rendition” of the Star-Spangled Banner at Woodstock when he performed it aloud in 1969… Music is neither recited, rendered, nor played when a recording (electronic or otherwise) is simply delivered to a potential listener.

In other words, a public performance in the internet context is precisely what one would think: a transmission of some sort in which the music encoded in the bits is being played back such that the encoded sounds are being reproduced by a computer. Or, as the Court put it, “Th[e] transmission, like a television or radio broadcast, is a performance because there is a playing of the song that is perceived simultaneously with the transmission.” Without a performance, there can be no public performance. At least the Court was having some fun in what was likely a tedious exposition of the statutory definitions.

Given this highly logical and fairly obvious interpretation of the statute, I find it hard to understand why ASCAP not only sued Yahoo and RealNetworks under this theory but chose to appeal the unfavorable decision to the Second Circuit. The only explanation I can come up with is that ASCAP was thinking about the royalties they receive from internet music streaming services and figured there shouldn’t be a distinction between a bit traveling across the network for one purpose versus any other. Maybe there is more to the case than is available publicly.

The Court also addressed the fee assessment made by the trial court. Because the Second Circuit remanded that issue and, moreover, because it is of much less import, I won’t discuss that portion of the Second Circuit’s opinion.

---

Meta: Apologies for the absence of activity on this blog of late. I have been too busy with client work and some other projects to devote any time to writing. But I did manage to find some time this week and fully intend to blog again on a regular basis.

The 9th Circuit Hammers Another Nail into the Coffin of the First Sale Doctrine

The Ninth Circuit’s recent decision in Vernor v. Autodesk has caused quite a stir among legal commentators and pundits. Denise Howell links to a few examples of the hysteria in her post on the case. In its decision last week, the Ninth Circuit Court of Appeals held that, where the software maker uses certain magic (and industry standard) words in its software license agreement, it can effectively prohibit its customers from reselling software, even software that is distributed on physical media and which the customer has uninstalled prior to resale. As I argue below, this should not come as a surprise to anyone.

First Sale. Normally, the purchaser of a book or a CD is entitled to resell the item without restriction, despite the otherwise exclusive rights granted by the Copyright Act to copyright owners to distribute their copyrighted works. This exception to the exclusive rights under copyright is known as the first sale doctrine and is codified at 17 U.S.C. §109.

In this case, Vernor purchased used copies of Autodesk’s software and posted it for resale on eBay. After eBay pulled Vernor’s listings, Vernor sued Autodesk for a declaratory judgment that his sales are authorized by the first sale doctrine. Because §109 only offers the benefit of the first sale doctrine to the “owner of a particular copy,” the crux of the issue came down to whether Vernor had title to his copies of the Autodesk software. In order to own the copies, there must have been a first sale by Autodesk of those copies, rather than a mere license.

Sale vs. License. In determining when a transaction is a “sale” versus a mere license, the Court articulated the following test. Does the copyright owner:

1. Characterize the transaction as a license?
2. Significantly restrict the user’s ability to transfer the software?
3. Impose notable use restrictions?

If the answer to all three is yes then it is a license and not a sale. According to the Court, Autodesk’s agreement satisfied all three prongs of the test. First, it used “license” and “licensee” language throughout. Second, the agreement stated that the software could not be transferred or leased without Autodesk’s consent. Third, it imposed restrictions on, inter alia, modifying or reverse engineering the software. Of course all of this is standard language for a software license agreement, and has been for as long as I have been practicing law. No IP licensing attorney worth her salt would draft an agreement without it.

I understand that part of the reason for all the chicken little hysteria are the potentially far-reaching effects of this decision. With this precedent in place, it may be possible for the owners of all sorts of copyrighted works—books, CDs, DVDs, etc.—to trump the first sale doctrine by attaching a license agreement to the works.

However, when it comes to digital works not distributed in physical form, last week’s decision does not change anything. The sky has already fallen for the those who wish to preserve the first sale doctrine in the digital age. It is already the case that software you cannot legally resell software you download, songs or movies you download, or ebooks you buy from Amazon and Apple. Whether the limitation is legal (a distribution is also a copy for strictly digital copies) or technological (DRM), the result is the same. With our move away from physical media, it appears that the first sale doctrine may be a relic of the 20th century mode of consuming creative works, and software.

Bona Fide Purchaser for Value Without Notice?

Vernor was aware of the terms of the license agreement, although there was probably not an offer and acceptance sufficient to establish formation of the contract. But an eBay buyer would likely not have been aware of the terms. So it is a strange result when a subsequent purchaser of software on physical media who has no privity with the licensor can be held liable for copyright infringement when loading the software onto her hard drive, or into RAM (if one does not own title to the copy, one cannot benefit from the essential step defense found in 17 U.S.C. §117(a)). What about this downstream purchaser, the eBay buyer who is not a licensing attorney and hasn’t read the 9th Circuit’s decision in Vernor v. Autodesk? It seems unrealistic and overly burdensome to require the buyer of copyright works on eBay to investigate the chain of title to the used media she is buying. What sort of liability would such an individual have in this case?

With respect to copyright ownership, and in certain personal property jurisprudence (e.g., negotiable instruments, promissory notes), there is a concept of the bona fide purchaser in good faith without notice, a buyer who has received possession of some copyrighted work or tangible property and has rights superior to a prior purchaser. (One benefit of recording copyright transfers is blocking application of this doctrine.) Should a similar doctrine apply here?

As the Court points out, whether its decision is the right result from an IP policy perspective is for Congress to decide. Given the relative lack of a circuit split here, the US Supreme Court is unlikely to hear this case. However, I understand that Vernor may appeal to an en banc panel of the Ninth Circuit. Perhaps first sale isn’t completely dead in the 9th Circuit.