Six interesting technology law issues raised in the Facebook IPO

Patent trolls, open source, do not track, SOPA, PIPA and much, much more: Facebook’s IPO filing has a real zoo of issues.

The securities laws require that companies going public identify risk factors that could adversely affect the company’s stock. Facebook’s S-1 filing, which it sent to the SEC today, identified almost 40 such factors. A number of these risks are examples of technology law issues that almost any internet company would face, particularly companies whose product is the users.

(1) Advertising regulation. In providing detail about the nature of this risk, Facebook mentions “adverse legal developments relating to advertising, including legislative and regulatory developments” and “the impact of new technologies that could block or obscure the display of our ads and other commercial content.” Facebook is likely concerned about the various technological and legal restrictions on online behavioral advertising, whether in the form of mandatory opportunities for users to opt-out of data collection or or the more aggressive “do not track” idea. The value of the advertising is of course tied to its effectiveness, and any technological, regulatory or legislative measures to enhance user privacy is a risk to Facebook’s revenue.

(2) Data security. No one knows exactly how much information Facebook has about its users. Not only does it have all the content uploaded by its 845 million users, it has the information that could be gleaned from the staggering 100 billion friendships among those users. [More stats] A data breach puts Facebook at risk of a PR backlash, regulatory investigations from the FTC, and civil liability to its users for negligence and other causes of action. But Facebook would not be left without remedy, having in its arsenal civil actions under the Computer Fraud and Abuse Act and the Stored Communications Act (among other laws) against the perpetrators. It is also likely the federal government would step in to enforce the criminal provisions of these acts as well.

(3) Changing laws. The section of the S-1 discussing this risk factor provides a laundry list of the various issues that online businesses face. Among them: user privacy, rights of publicity, data protection, intellectual property, electronic contracts, competition, protection of minors, consumer protection, taxation, and online payment services. Facebook is understandably concerned that changes to any of these areas of the law, anywhere in the world, could make doing business more expensive or, even worse, make parts of the service unlawful. Though not mentioned by name here, SOPA, PIPA, and do-not-track legislation are clearly in Facebook’s mind when it notes that “there have been a number of recent legislative proposals in the United States . . . that would impose new obligations in areas such as privacy and liability for copyright infringement by third parties.”

(4) Intellectual property protection. The company begins its discussion of this risk with a few obvious observations, namely, how the company may be adversely affected if it is unable to secure trademark, copyright or patent registration for its various intellectual property assets. Later in the disclosure, though, Facebook says some really interesting things about open source:

As a result of our open source contributions and the use of open source in our products, we may license or be required to license innovations that turn out to be material to our business and may also be exposed to increased litigation risk. If the protection of our proprietary rights is inadequate to prevent unauthorized use or appropriation by third parties, the value of our brand and other intangible assets may be diminished and competitors may be able to more effectively mimic our service and methods of operations.

(5) Patent troll lawsuits. Facebook notes that internet and technology companies “frequently enter into litigation based on allegations of infringement, misappropriation, or other violations of intellectual property or other rights.” But it goes on to give special attention to those “non-practicing entities” (read: patent trolls) “that own patents and other intellectual property rights,” which “often attempt to aggressively assert their rights in order to extract value from technology companies.” Facebook believes that as its profile continues to rise, especially in the glory of its IPO, it will increasingly become the target of patent trolls. For now it does not seem worried: “[W]e do not believe that the final outcome of intellectual property claims that we currently face will have a material adverse effect on our business.” Instead, those endeavors are a suck on resources: “[D]efending patent and other intellectual property claims is costly and can impose a significant burden on management and employees….” And there is also the risk that these lawsuits might turn out badly, and Facebook would have to pay judgments, get licenses, or develop workarounds.

(6) Tort liability for user-generated content. Facebook acknowledges that it faces, and will face, claims relating to information that is published or made available on the site by its users, including claims concerning defamation, intellectual property rights, rights of publicity and privacy, and personal injury torts. Though it does not specifically mention the robust immunity from liability over third party content provided by 47 U.S.C. 230, Facebook indicates a certain confidence in the protections afforded by U.S. law from tort liability. It is the international scene that gives Facebook concern here: “This risk is enhanced in certain jurisdictions outside the United States where our protection from liability for third-party actions may be unclear and where we may be less protected under local laws than we are in the United States.”

You have to hand it to the teams of professionals who have put together Facebook’s IPO filing. I suppose the billions of dollars at stake can serve as a motivation for thoroughness. In any event, the well-articulated discussion of these risks in the S-1 is an interesting read, and can serve to guide the many lesser-valued companies out there.

Fair use, the DMCA, and presidential politics

The 2012 presidential election cycle is already giving internet law enthusiasts things to talk about. Last week it was Ron Paul’s grumblings about an unauthorized campaign ad on YouTube. Now NBC is moaning about a Mitt Romney ad comprised almost entirely of Tom Brokaw on the Nightly News in 1997.

NBC has asked the ad be pulled, claiming it is a copyright infringement. Smart people are already saying the ad is fair use. It probably is fair use.

And NBC knows that. Romney’s campaign posted the ad on YouTube five days ago, and it is yet to be the subject of a DMCA takedown notice. Though such a notice would be easy to draft and send, NBC is aware that the fallout could be expensive. Section 512(f) of the DMCA penalizes the senders of bogus takedown notices. And the courts have not taken kindly to purported victims of infringement who do not fully consider fair use before having content taken off YouTube.

With the election still months away, we may yet see controversial action like we did in 2008 by the news media to disable political content. These situations underscore the problem presented by how long it takes to process DMCA counternotifications and 512(f) actions.

A candidate’s defeat makes these processes moot. So maybe we should hope for a longer republican primary season just so we can see some good DMCA and fair use litigation. Come on NBC, send that takedown notice!

Are nonpirate Megaupload users entitled to compensation from the government?

If I left my coat in a taxi that was later impounded because, unknown to me, the driver was transporting heroin in the trunk, would I be left out in the cold?

People who used Megaupload to lawfully store and transfer files are rightfully upset that their stuff is unavailable after last week’s raid. Some groups in other countries say they are going to sue the U.S. government. Would a lawsuit like that get anywhere in a U.S. court?

The Fifth Amendment — best known for its privilege against self-incrimination — says that “private property [shall not] be taken for public use, without just compensation”. (You can impress your legally-trained friends at parties by confidently and casually referring to the Takings Clause.) Does the Takings Clause give innocent Megaupload users a right to be paid the value of the files they are being deprived of while the feds use the servers on which those files are stored to prove their case against Kim Dotcom and company?

Back in 2008, Ilya Somin and Orin Kerr had a conversation on the Volokh Conspiracy discussing this question of whether the Fifth Amendment protects innocent third parties who lose property in a criminal investigation. If you read that commentary you will see that a case over the Megaupload takedown might be tough for a number of esoteric reasons, not the least of which is Supreme Court precedent.

There are some face-value problems with a case like this as well. Has the government taken the property for a “public use”? One could argue that the reason the servers (including the innocent content) were seized was for the so-called public good of going after piracy. But then the innocent content is not being “used” in connection with the prosecution — it just happens to be there.

I do not pretend to know the answers to this inquiry, and I’m relying on sharper Constitutional minds than mine to leave some good comments. (If you know Ilya Somin or Orin Kerr, send them a link to this post!) All I know is that it does not seem fair that users of the cloud should so easily be deprived in the name of law enforcement.

 

Enhanced by Zemanta

If you critique SOPA, read the text. If you read the text, read it right.

Earlier this week Eriq Gardner speculated in a tweet that less than one tenth of one percent of folks have actually read the SOPA legislation. I bet he’s right. It’s good to read the statute. But what might be worse than not reading it is reading it wrongly and thereafter propagating misunderstanding.

One of the motifs that has permeated the SOPA discussion is this idea that evil (usually corporate) interests could shut down entire, innocent sites based on one piece of user generated content on that site that is, or links to, infringing material.

Some commentators, such as the usually astute Khan Academy in the video embedded below, have gone so far as to say that one little transgression by one user could bring down all of Facebook, YouTube, or Vimeo. (That discussion begins at about the 5:00 mark where the narrator purports to parse the language of Section 103 of SOPA.)

We are fortunate to have the means and motivation to rally around an issue like SOPA and make it a topic of worldwide discussion. But it turns unfortunate when some of the loudest criticism is based on misinformation. That’s happening now.

It is silly to think that one person could bring down Facebook and leave its almost a billion users in the dark. It is silly to think that Congress would enact legislation making that possible. Those thoughts are silly because they are not based on reality. One user could not cause Facebook to be shut down, and SOPA does provide content owners with a way to accomplish that.

If you take a close look at SOPA, (and of course you should) you see that this “one person taking down Facebook” conclusion is not supported by the language of the statute.

If a federal judge is convinced that a site is “dedicated to the theft of U.S. property,” then he or she can enter an injunction (according to the Federal Rules of Civil Procedure and all the case law attendant thereto) shutting down that site’s domain name.

The present misunderstanding comes from a reading of SOPA’s definition in Section 103 of what it means for a site to be “dedicated to the theft of U.S. property.” That definition is much narrower than what other commentators would have you believe. Among other things, the site has to be:

  • primarily designed or operated for the purpose of offering goods or services in a manner that engages in, enables, or facilitates infringement, circumvention or counterfeiting,
  • have only limited purpose or use other than offering goods or services in a manner that engages in, enables, or facilitates infringement, circumvention or counterfeiting, or
  • be marketed by its operator or another acting in concert with that operator for use in offering goods or services in a manner that engages in, enables, or facilitates infringement, circumvention or counterfeiting.

A less-than-careful reading leads one to think that the definition brings in any site that enables or facilitates infringement, circumvention or counterfeiting. Read the definition again. Is Facebook primarily designed to rip off US property? Is it used for only a little more (i.e., does it have a limited purpose) other than to enable or promote piracy? Does Mark Zuckerberg say that is what it is for? Because the answers to these questions are no, no and no, a federal judge would not conclude that Facebook is a site dedicated to the theft of U.S. property.

If that federal judge were to so conclude, then he would likely be smoking dope. And if that is the state of affairs, then our problem is not SOPA, but federal judges smoking dope.

The fervent opposition to SOPA leads one to be reminded, like David Newhoff was in this piece, of the “death panel” hyperbole that surrounded the healthcare debate. It might be the same part of the brain at work that caused all the irrationality in Vancouver after the Stanley Cup. I’m just sayin’.

Principled and reasonable debate is awesome. Misguided arguments waste everyone’s time.

No deposition of account holder allowed until he is named as defendant in BitTorrent copyright case

Hard Drive Productions, Inc. v. Doe, 2012 WL 90412 (E.D. Cal. July 11, 2012)

In a mass copyright infringement suit, plaintiff served a subpoena on an internet service provider and got the identifying information for the account holder suspected of trading a copy of a movie via BitTorrent. The account holder was uncooperative with plaintiff’s offers to settle, and denied downloading the file.

Instead of simply naming the identified account holder as a defendant in the case and proceeding with ordinary discovery, plaintiff asked the court for leave to take “expedited discovery,” namely, to depose the account holder to learn about:

  • the account holder’s involvement with the alleged distribution
  • his computers and network setup
  • his technical savvy
  • other users who may have had access to the computers or network

The court denied plaintiff’s request for leave to engage in the expedited discovery. It found that unlike other copyright cases in which anonymous infringers were identified, the efforts in this case “went far beyond seeking to identify a Doe defendant.” Instead, the court observed, it would be “a full-on deposition during which [the account holder] who plaintiff admits is likely not represented by counsel, may unwarily incriminate himself on the record before he has even been named as a defendant and served with process.”

Video: my appearance on the news talking about isanyoneup.com

Last night I appeared in a piece that aired on the 9 o’clock news here in Chicago, talking about the legal issues surrounding isanyoneup.com. (That site is definitely NSFW and I’m not linking to it because it doesn’t deserve the page rank help.) The site presents some interesting legal questions, like whether and to what extent it is shielded by Section 230 of the Communications Decency Act for the harm that arises from the content it publishes (I don’t think it is shielded completely). The site also engages in some pretty blatant copyright infringement, and does not enjoy safe harbor protection under the Digital Millennium Copyright Act.

Here’s the video:

Ninth Circuit: Apple did not engage in copyright misuse by restricting OS X to Apple hardware

Apple Inc. v. Psystar Corp., — F.3d —, 2011 WL 4470623 (9th Cir. September 28, 2011) [PDF]

Back in 2008, Apple sued Psystar for copyright infringement arising from Psystar’s manufacture and distribution of computers preloaded with copies of Mac OS X. Psystar lost at the trial court level, with the judge rejecting its argument that Apple engaged in anti-competitive, “copyright misuse” by requiring in its OS X software license agreement that the operating system be used only on Apple hardware. Psystar sought review of this ruling. On appeal, the Ninth Circuit affirmed.

Copyright misuse is a defense (not an independent cause of action) that one sued for infringement can raise. Courts will find that a plaintiff has engaged in copyright misuse if the enforcement of the plaintiff’s copyright will restrain the development of competing products. In this case, Psystar claimed that Apple’s enforcement of its software license restrained the development of competing hardware.

The court rejected that argument because Apple’s enforcement of its software license agreement, requiring that the software be used only on Apple hardware, did not restrict Psystar from developing its own software. The court found that:

Apple’s [software license agreement] does not restrict competitor’s [sic.] to develop their own software, nor does it preclude customers from using non-Apple components with Apple computers. Instead, Apple’s [software license agreement] merely restricts the use of Apple’s own software to its own hardware. . . . Psystar produces its own computer hardware and it is free to develop its own computer software.

This case solidifies Apple’s approach to enforcing a controlled, closed ecosystem for the distribution of software used for Macs and iDevices. Now that a federal court has found that Apple is not playing unfairly by keeping its users from loading Apple software onto non-Apple hardware, the company can likewise maintain the technological controls that ensure only approved applications are used in connection with the operating systems. Marketplaces for third-party hardware running Apple software would greatly lower the entry barrier for hackers and enthusiasts to play outside of the rules. But this decision from the Ninth Circuit keeps those rules firmly in place.

Court deals blow to anonymous Bittorrent defendants’ efforts to challenge subpoenas

West Coast Productions v. Does 1 – 5,829, — F.Supp.2d —, 2011 WL 2292239 (D.D.C. June 10, 2011)

The judge in one of the well-known mass copyright cases filed by Dunlap, Grubb & Weaver a/k/a U.S. Copyright Group (West Coast Productions v. Does 1 – 5,829) has issued an order denying motions to quash filed by several of the unnamed defendants. Plaintiff had served subpoenas on the ISPs associated with the IP addresses allegedly involved in Bittorrent activity, seeking to learn the identity of those account holders.

The ruling is potentially troubling because the court refused to even consider the arguments presented by those anonymous parties who did not reveal their identity in connection with the motion to quash. Such an approach undermines, and indeed comes close to refusing altogether to recognize any privacy interest that a person may have concerning his or her ISP account information.

The court observed that the Federal Rules of Civil Procedure require that a party must identify himself or herself in the papers filed with the court. In some situations, however, a court may grant a “rare dispensation” of anonymity after taking into account the risk of unfairness to the party seeking anonymity as well as the general presumption of openness in judicial proceedings.

In this case, the court noted that other courts had “uniformly held that the privacy interest in [ISP account] information is minimal and not significant enough to warrant the special dispensation of anonymous filing.”

Absent from the court’s analysis was the potential for harm to defendants who were the subject of these subpoenas but might have the ability to demonstrate (anonymously) that they were not involved. In cases involving adult content, in particular, the harm of being publicly associated with that content — even if the association turns out to be in error — is one that should not be disregarded in this way. Moreover, taking away the ability of an anonymous defendant to challenge his unmasking will encourage extortionate-like behavior on the part of copyright plaintiffs hoping to extract a settlement early in the case. If writing a check is the only way to keep from having to turn one’s name over (and this case pretty much establishes that rule), then more settlements should be expected.

The court went on to reject the arguments in favor of motions to quash filed by John Does who had provided their contact information to the court. The court found that it was premature to rule on any objections based on a lack of personal jurisdiction because the defendants filing the motions had not actually been named as a party. And the court rejected the arguments that the defendants were improperly joined into the action, noting the allegations in the complaint that the IP addresses were involved in a single Bittorrent swarm.

Evan Brown is a Chicago-based attorney practicing technology and intellectual property law. Send email to ebrown@internetcases.com, call (630) 362-7237, follow on Twitter at @internetcases, and be sure to like Internet Cases on Facebook.

More subpoenas on the way to identify John Doe BitTorrent users in copyright cases

First Time Videos v. Does 1-37, 2011 WL 1431619 (N.D. California, April 14, 2011)

Hard Drive Productions v. Does 1-118, 2011 WL 1431612 (N.D. California, April 14, 2011)

There have been a couple of new cases filed in federal court in California alleging that unknown BitTorrent users committed copyright infringement and engaged in civil conspiracy by trading porn files online. [Read about some earlier, ongoing cases of this type here and here]. The court has issued orders that move the process of uncovering the identities of the John Doe defendant BitTorrent users.

Generally a plaintiff cannot start the discovery process in a case until it has had a “Rule 26(f)” conference with the defendant. But when the defendants are anonymous (as they are in these BitTorrent cases — they’re known only by IP address), the plaintiff has a bit of a problem. It needs discovery to find out the names of the defendants, but it cannot take discovery before the Rule 26(f) conference. [More on this]

So in cases like this, a plaintiff will ask the court to allow the early discovery to be had. Courts grant those motions allowing early discovery when good cause has been shown.

In this case, the court allowed the discovery because the following four criteria had been met:

(1) The plaintiffs had identified the Doe defendants with sufficient specificity that the court could determine that the defendants are real people who can be sued in federal court. On this point, the court credited the list of IP addresses associated with each of the unknown defendants.

(2) The plaintiffs recounted the steps taken to locate and identify the defendants. Again, the court looked to the fact that the defendants were known only by IP addresses. The names of the defendants could not be ascertained from the information available.

(3) The plaintiffs demonstrated that the action could withstand a motion to dismiss. In some cases this is a tough hurdle to get over. But in copyright cases the threshold can be met relatively easily — simply alleging ownership of a copyright and unlawful copying satisfies this element.

(4) The plaintiffs proved that the discovery was likely to lead to identifying information that will permit service of process. Getting the subscriber information from the ISPs would allow names to be associated with the IP addresses, for further action to be taken.

(The above 4-factor test is drawn from Columbia Ins. Co. v. seescandy. com, 185 F.R.D. 573, 578–80 (N.D.Cal.1999).)

So ISPs across the country will be getting peppered with more subpoenas, and sending out letters to their John Doe subscribers, giving deadlines to move to quash the subpoenas. More mad scramble to protect identities is on its way.

Court leaves thousands of BitTorrent copyright infringement defendants joined in single action

Call of the Wild Movie v. Does 1 – 1,062 — F.Supp.2d —, 2011 WL 996786 (D.D.C. March 22, 2011)

One of the craziest things about the copyright infringement lawsuits that have been brought against BitTorrent users accused of trading movies over the internet is the vast number of John Doe defendants that are usually lumped into one case. After the plaintiff copyright owners file a complaint for infringement — sometimes against thousands of anonymous defendants — they ask the court for leave to take expedited discovery. Then the movie companies serve subpoenas on the John Does’ internet service providers, asking the ISPs to disclose the identities of their customers associated with particular IP addresses.

Prosecuting a case against thousands of copyright infringement defendants is an enormous task, both for the plaintiffs’ attorneys as well as the ISPs who must respond to the subpoenas. Having so many defendants risks making the case unmanageable. So one may question whether it is appropriate under the Federal Rules of Civil Procedure to have so many unknown defendants all in the same case. In the nomenclature of civil litigation, the question is whether the joinder of all the defendants in one action is appropriate.

In three of the BitTorrent copyright cases pending in federal court in Washington DC brought by the US Copyright Group on behalf of a handful of independent film makers, groups like the Electronic Frontier Foundation, the ACLU and others argued improper joinder. These organizations filed amicus briefs in the cases of Call of the Wild Movie v. Does 1 – 1,062, Maverick Entertainment v. Does 1 – 4,350, and Donkeyball Movie v. Does 1 – 171, arguing that joining all the defendants in one action violated Rule 20 of the Federal Rules of Civil Procedure. The court rejected these arguments, finding that joinder was proper, at least in such early stages of the litigation where the defendants had not yet been identified.

The court considered three factors when answering the question of proper joinder: (1) whether the claims arose from the same transaction or occurrence or series of transactions or occurrences, (2) whether the legal and factual questions are common to all defendants, (3) and whether joinder would cause prejudice to any party or needless delay.

Same transaction or occurrence

The court observed that claims against joined parties must be “logically related,” and that this is a flexible test, with courts seeking the broadest possible scope of action. The court held that the claims against the BitTorrent users were logically related, based on plaintiffs’ allegations that the BitTorrent protocol makes every downloader of a file also an uploader, and accordingly, every user who has a copy of the infringing file on the network must necessarily be a source of download for that infringing file. This is an interesting finding, in that the strength of plaintiffs’ allegations were based on how BitTorrent works.

Common legal and factual questions

As for this second factor, the court found that the legal and factual questions were common because the parties would be litigating the same copyright claims, and all of the claims related to the use of BitTorrent.

Prejudice or needless delay

The court said some intriguing things about the interests of the parties in making its findings on this factor. For one, it said that leaving all the defendants joined in the same action would benefit them all, in that they would be able to see the defenses that other defendants were making. The court also expressed concern in favor of the efficiencies afforded the plaintiffs in filing these mass lawsuits. The plaintiff movie studios have been criticized for filing suit against large numbers of defendants in one action rather than separate suits against each defendant (and thereby having to pay only one filing fee to start the action versus several thousand filing fees). The court saw this question squarely in favor of plaintiff. It found that forcing plaintiffs to administer multiple actions, and having to pay the filing fees in all those actions “would certainly not be in the ‘interests of convenience and judicial economy,’ or ‘secure the just, speedy and inexpensive determination of the action.'”

Scroll to top