The Technology Policy Institute has posted the video of my talk at the 2024 Aspen Forum panel on “How Should we Regulate the Digital World?” My remarks run from 33:33–44:12 of the video. I also elaborate briefly during Q&A.

My remarks at this year’s TPI Aspen Forum panel were derived from my R Street Institute essay, “The Policy Origins of the Digital Revolution & the Continuing Case for the Freedom to Innovate,” which sketches out a pro-freedom vision for the Computational Revolution.

Over at Discourse magazine this week, my R Street colleague Jonathan Cannon and I have posted a new essay on how it has been “Quite a Fall for Digital Tech.” We mean that both in the sense that the last few months have witnessed serious market turmoil for some of America’s leading tech companies, but also that the political situation for digital tech more generally has become perilous. Plenty of people on the Left and the Right now want a pound of flesh from the info-tech sector, and the starting cut at the body involves Section 230, the 1996 law that shields digital platforms from liability for content posted by third parties.

With the Supreme Court recently announcing it will hear Gonzalez v. Google, a case that could significantly narrow the scope of Section 230, the stakes have grown higher. It was already the case that federal and state lawmakers were looking to chip away at Sec. 230’s protections through an endless variety of regulatory measures. But if the Court guts Sec. 230 in Gonzalez, then it will really be open season on tech companies, as lawsuits will fly at every juncture whenever someone does not like a particular content moderation decision. Cannon and I note in our new essay that,

if the court moves to weaken liability protections for digital platforms, the ramifications will be profoundly negative. While many critics today complain that the law’s liability protections have been too generous, the reality is that Section 230 has been the legal linchpin supporting the permissionless innovation model that fueled America’s commanding lead in the digital information revolution. Thanks to the law, digital entrepreneurs have been free to launch bold new ideas without fear of punishing lawsuits or regulatory shenanigans. This has boosted economic growth and dramatically broadened consumer information and communications options.

Many critics of Sec. 230 claim that reforms are needed to “rein in Big Tech.” But, ironically, gutting Sec. 230 would probably only make big tech companies even bigger because the smaller players in the market would struggle to deal with the mountains of regulations and lawsuits that would come about in its absence. Cannon and I continue on to explore what it means for the next generation of online innovators if these court cases go badly and Section 230 is scaled back or gutted:

Section 230 has been a legal cornerstone of the entire ecosystem. All the large-scale platforms we depend on for our online experience would never have gotten off the ground without its protection. […] More importantly, these platforms have relied on being able to host third-party content without fear of opening a Pandora’s box of private litigation and endless challenges from governments. By removing these protections, platforms will be forced to significantly increase their moderation practices to reduce risk of suits from zealous litigants. Besides the chilling effect this will have on speech, it also will put up a cost-prohibitive barrier for smaller entrants who lack the resources to have an army of content moderators to find and eliminate undesirable content.

The broader effect on market dynamism and the nation’s technological competitiveness will be profound as permissionless innovation is replaced by mountains of top-down permission slips. “If America’s digital sector gets kneecapped by the Supreme Court, or if new regulations or legislative proposals scale back Section 230 protections, it will be significantly more difficult for U.S. firms to continue to lead in the development and commercialization of new technologies,” we conclude.

Jump over to Discourse to read the entire piece.

My colleague Wayne Brough and I recently went on the “Kibbe on Liberty” show to discuss how to discuss the state of free speech on the internet. We explained how censorship is a Big Government problem, not a Big Tech problem. Here’s the complete description of the show and the link to the full episode is below.

” With Elon Musk’s purchase of Twitter, we are in the middle of a national debate about the tension between censorship and free expression online. On the Right, many people are calling for government to rein in what they perceive as the excesses of Big Tech companies, while the Left wants the government to crack down on speech they deem dangerous. Both approaches make the same mistake of giving politicians authority over what we are allowed to say and hear. And with recent revelations about government agents leaning on social media companies to censor speech, it’s clear that when it comes to the online conversation, there’s no such thing as a purely private company.”

For more on this issues, please see: “The Classical Liberal Approach to Digital Media Free Speech Issues.”

By: Jennifer Huddleston and Juan Martin Londoño

This year the E3 conference streamed live over Twitch, YouTube, and other online platforms—a reality that highlights the growing importance of platforms and user-generated content to the gaming industry. From streaming content on Twitch, to sharing mods on Steam Workshop, or funding small developing studios on services such as Patreon or Kickstarter, user-generated content has proven vital for the gaming ecosystem. While these platforms have allowed space for creative interaction—which we saw on the livestreams chats during E3—the legal framework that allows all of this interaction is under threat, and changes to a critical internet law could spell Game Over for user-created gaming elements.

This law, “Section 230,” is foundational to all user-generated content on the internet. Section 230 protects platforms from lawsuits over both the content they host as well as their moderation decisions, giving them the freedom to curate and create the kind of environment that best fits its customers. This policy is under attack, however, from policymakers on both sides of the aisle. Some Democrats argue platforms are not moderating enough content, thus allowing hate speech and voter suppression to thrive, while some Republicans believe platforms are moderating too much, which promotes “cancel culture” and the limitation of free speech.

User-generated content and the platforms that host it have contributed significantly to the growth of the gaming industry since the early days of the internet. This growth has only accelerated during the pandemic, as in 2020 the gaming industry grew 20 percent to a whopping $180 billion market. But changing Section 230 could seriously disrupt user-generated engagement with gaming, making content moderation costlier and riskier for some of gamers’ favorite platforms.

An increased legal liability could mean a platform such as Twitch would face higher compliance costs due to the need to increase its moderation and legal teams. This cost would likely be transferred to creators through a revenue reduction or to viewers through rate hikes—resulting in less content and fewer users. Further, restrictions on moderation could lead to undesirable content and ultimately fewer users and advertisers—leading to more profit losses and less content. Ultimately, platforms might not be able to sustain themselves, leading to fewer platforms and opportunities for fans to engage. Platforms such as Twitch already face these problems, but for now they can determine the best solutions without heavy-handed government intervention or costly legal battles.

The impact of changing Section 230 goes beyond video content and could impact some increasingly popular fan creations that are further invigorating the industry. For example, the modding community, composed of gaming fans that modify existing games to create new experiences, often uses various online platforms to share their mods with other players. Modding has kept certain games relevant even years after their release, or propelled games’ popularity by introducing new ways to play them. Such is the case of Grand Theft Auto V’s roleplaying mod, or Arma III’s PlayerUnknown Battlegrounds mod, the inspiration of games such as Fortnite and Call of Duty: Warzone.

These modified games are often hosted on platforms such as Steam Workshop, Github, or on independently run community websites. These platforms are often free of charge, either as a complimentary service of a bigger product – in the case of Steam – or are supported purely by ad revenue and donations. Like streaming platforms and message boards, without Section 230 these services would face increased compliance costs or be unable to remove excessively violent, sexually explicit, or hateful content. The result could be that these new twists on old favorites never make it to consumers, as platforms are unable to host these creations and remain viable as businesses.

Changing or removing Section 230 protections would upend the complex and dynamic gaming environment on display during E3. It took decades of growth for gaming to establish itself as the new king of entertainment and it has defended itself from a variety of technopanics throughout the years. Pulling the plug on Section 230 could mean “Game Over” for the user-generated content that brings gamers so much fun.

Over at Discourse magazine I’ve posted my latest essay on how conservatives are increasingly flirting with the idea of greatly expanding regulatory control of private speech platforms via some sort of common carriage regulation or new Fairness Doctrine for the internet. It begins:

Conservatives have traditionally viewed the administrative state with suspicion and worried about their values and policy prescriptions getting a fair shake within regulatory bureaucracies. This makes their newfound embrace of common carriage regulation and media access theory (i.e., the notion that government should act to force access to private media platforms because they provide an essential public service) somewhat confusing. Recent opinions from Supreme Court Justice Clarence Thomas as well as various comments and proposals of Sen. Josh Hawley and former President Trump signal a remarkable openness to greater administrative control of private speech platforms. Given the takedown actions some large tech companies have employed recently against some conservative leaders and viewpoints, the frustration of many on the right is understandable. But why would conservatives think they are going to get a better shake from state-regulated monopolists than they would from today’s constellation of players or, more importantly, from a future market with other players and platforms?

I continue on to explain why conservatives should be skeptical of the administrative state being their friend when it comes to the control of free speech. I end by reminding conservatives what President Ronald Reagan said in his 1987 veto of legislation to reestablish the Fairness Doctrine: “History has shown that the dangers of an overly timid or biased press cannot be averted through bureaucratic regulation, but only through the freedom and competition that the First Amendment sought to guarantee.”

Read more at Discourse, and down below you will find several other recent essays I’ve written on the topic.

• “FCC’s O’Rielly on First Amendment & Fairness Doctrine Dangers“
• “A Good Time to Re-Read Reagan’s Fairness Doctrine Veto“
• “Sen. Hawley’s Radical, Paternalistic Plan to Remake the Internet“
• “How Conservatives Came to Favor the Fairness Doctrine & Net Neutrality“
• “Sen. Hawley’s Moral Panic Over Social Media“
• “The White House Social Media Summit and the Return of ‘Regulation by Raised Eyebrow’“
• “The Not-So-SMART Act“
• “The Surprising Ideological Origins of Trump’s Communications Collectivism“

In a five-part series at the American Action Forum, I presented prior to the 2020 presidential election the candidates’ positions on a range of tech policy topics including: the race to 5G, Section 230, antitrust, and the sharing economy. Now that the election is over, it is time to examine what topics in tech policy will gain more attention and how the debate around various tech policy issues may change. In no particular order, here are five key tech policy issues to be aware of heading into a new administration and a new Congress. 

The  Use of Soft Law for Tech Policy 

In 2021, it is likely America will still have a divided government with Democrats controlling the White House and House of Representatives and Republicans expected to narrowly control the Senate. The result of a divided government, particularly between the two houses of Congress, will likely be that many tech policy proposals face logjams. The result will likely be that many of the questions of tech policy lack the legislation or hard law framework that might be desired. As a result, we are likely to continue to see “soft law”—regulation by various sub-regulatory means such as guidance documents, workshops, and industry consultations—rather than formal action. While it appears we will see more formal regulatory action from the administrative state as well in a Biden Administration, these actions require quite a process through comments and formal or informal rulemaking. As technology continues to accelerate, many agencies turn to soft law to avoid “pacing problems” where policy cannot react as quickly as technology and rules may be outdated by the time they go into effect. 

A soft law approach can be preferable to a hard law approach as it is often able to better adapt to rapidly changing technologies. Policymakers in this new administration, however, should work to ensure that they are using this tool in a way that enables innovation and that appropriate safeguards ensure that these actions do not become a crushing regulatory burden. 

Return of  the  Net Neutrality  Debate 

One key difference between President Trump and President-elect Biden’s stances on tech policy concerns whether the Federal Communication Commission (FCC) should categorize internet service providers (ISPs) as Title II “common carrier services,” thereby enabling regulations such as “net neutrality” that places additional requirements on how these service providers can prioritize data. President-elect Biden has been clear in the past that he favors reinstating net neutrality. 

The imposition of this classification and regulations occurred during the Obama Administration and the FCC removed both the classification under Title II and the additional regulations for “net neutrality” during the Trump Administration. Critics of these changes made many hyperbolic claims at the time such as that Netflix would be interrupted or that ISPs would use the freedom in a world without net neutrality to block abortion resources or pro-feminist groups. These concerns have proven to be misguided. If anything, the COVID-19 pandemic has shown the benefits to building a robust internet infrastructure and expanded investment that a light-touch approach has yielded. 

It is likely that net neutrality will once again be debated. Beyond just the imposition of these restrictions, a repeated change in such a key classification could create additional regulatory uncertainty and deter or delay investment and innovation in this valuable infrastructure. To overcome such concerns, congressional action could help fashion certainty in a bipartisan and balanced way to avoid a back-and-forth of such a dramatic nature. 

Debates Regarding  Sharing Economy Providers ’   Classification  as Independent Contractors 

California voters passed Proposition 22 undoing the misguided reclassification of app-based service drivers as employees rather than independent contractors under AB5; during the campaign, however, President-elect Biden stated that he supports AB5 and called for a similar approach nationwide. Such an approach would make it more difficult on new sharing economy platforms and a wide range of independent workers (such as freelance journalists) at a time when the country is trying to recover economically.  

Changing classifications to make it more difficult to consider service providers as independent contractors makes it less likely that platforms such as Fiverr or TaskRabbit could provide platforms for individuals to offer their skills. This reclassification as employees also misunderstands the ways in which many people choose to engage in gig economy work and the advantages that flexibility has. As my AAF colleague Isabel Soto notes, the national costs of a similar approach found in the Protecting the Right to Organize (PRO) Act “could see between $3.6 billion and $12.1 billion in additional costs to businesses” at a time when many are seeking to recover during the recession. Instead, both parties should look for solutions that continue to allow the benefits of the flexible arrangements that many seek in such work, while allowing for creative solutions and opportunities for businesses that wish to provide additional benefits to workers without risking reclassification. 

Shifting Conversations and Debates Around Section 230 

Section 230 has recently faced most of its criticism from Republicans regarding allegations of anti-conservative bias. President-elect Biden, however, has also called to revoke Section 230 and to set up a taskforce regarding “Online Harassment and Abuse.” While this may seem like a positive step to resolving concerns about online content, it could also open the door to government intervention in speech that is not widely agreed upon and chip away at the liability protection for content moderation. 

For example, even though the Stop Enabling Sex Trafficking Act was targeting the heinous crime of sex trafficking (which was already not subject to Section 230 protection) was aimed at companies such as Backpage where it was known such illegal activity was being conducted, it has resulted in legitimate speech such as Craigslist personal ads being removed  and companies such as Salesforce being subjected to lawsuits for what third parties used their product for. A carveout for hate speech or misinformation would only pose more difficulties for many businesses. These terms to do not have clearly agreed-upon meanings and often require far more nuanced understanding for content moderation decisions. To enforce changes that limit online speech even on distasteful and hateful language in the United States would dramatically change the interpretation of the First Amendment that has ruled such speech is still protected and would result in significant intrusion by the government for it to be truly enforced. For example, in the UK, an average of nine people a day were questioned or arrested over offensive or harassing “trolling” in online posts, messages, or forums under a law targeting online harassment and abuse such as what the taskforce would be expected to consider. 

Online speech has provided new ways to connect, and Section 230 keeps the barriers to entry low. It is fair to be concerned about the impact of negative behavior, but policymakers should also recognize the impact that online spaces have had on allowing marginalized communities to connect and be concerned about the unintended consequences changes to Section 230 could have. 

Continued Antitrust Scrutiny of “Big Tech” 

One part of the “techlash” that shows no sign of diminishing in the new administration or new Congress is using antitrust to go after “Big Tech.” While it remains to be seen if the Biden Department of Justice will continue the current case against Google, there are indications that they and congressional Democrats will continue to go after these successful companies with creative theories of harm that do not reflect the current standards in antitrust. 

Instead of assuming a large and popular company automatically merits competition scrutiny  or attempting to utilize antitrust to achieve policy changes for which it is an ill-fitted tool, the next administration should return to the principled approach of the consumer welfare standard. Under such an approach, antitrust is focused on consumers and not competitors. In this regard, companies would need to be shown to be dominant in their market, abusing that dominance in some ways, and harming consumers. This approach also provides an objective standard that lets companies and consumers know how actions will be considered under competition law. With what is publicly known, the proposed cases against the large tech companies fail at least one element of this test. 

There will likely be a shift in some of the claimed harms, but unfortunately scrutiny of large tech companies and calls to change antitrust laws to go after these companies are likely to continue. 

Conclusion 

There are many other technology and innovation issues the next administration and Congress will see. These include not only the issues mentioned above, but emerging technologies like 5G, the Internet of Things, and autonomous vehicles. Other issues such as the digital divide provide an opportunity for policymakers on both sides of the aisle to come together and have a beneficial impact and think of creative and adaptable solutions. Hopefully, the Biden Administration and the new Congress will continue a light-touch approach that allows entrepreneurs to engage with innovative ideas and continues American leadership in the technology sector. 

Last week I attended the Section 230 cage match workshop at the DOJ. It was a packed house, likely because AG Bill Barr gave opening remarks. It was fortuitous timing for me: my article with Jennifer Huddleston, The Erosion of Publisher Liability in American Law, Section 230, and the Future of Online Curation, was published 24 hours before the workshop by the Oklahoma Law Review.

These were my impressions of the event:

I thought it was pretty well balanced event and surprisingly civil for such a contentious topic. There were strong Section 230 defenders and strong Section 230 critics, and several who fell in between. There were a couple cheers after a few pointed statements from panelists, but the audience didn’t seem to fall on one side or the other. I’ll add that my friend and co-blogger Neil Chilson gave an impressive presentation about how Section 230 helped make the “long tail” of beneficial Internet-based communities possible.

AG Bob Barr gave the opening remarks, which are available online. A few things jumped out. He suggested that Section 230 had its place but Internet companies are not an infant industry anymore. In his view, the courts have expanded Section 230 beyond drafters’ intent, and the Reno decision “unbalanced” the protections, which were intended to protect minors. The gist of his statement was that the law needs to be “recalibrated.”

Each of these points were disputed by one or more panelists, but the message to the Internet industry was clear: the USDOJ is scrutinizing industry concentration and its relationship to illegal and antisocial online content.

The workshop signals that there is now a large, bipartisan coalition that would like to see Section 230 “recalibrated.” The problem for this coalition is that they don’t agree on what types of content providers should be liable for and they are often at cross-purposes. The problematic content ranges from sex trafficking, to stalkers, to opiate trafficking, to revenge porn, to unfair political ads. For conservatives, social media companies take down too much content, intentionally helping progressives. For progressives, social media companies leave up too much content, unwittingly helping conservatives.

I’ve yet to hear a convincing way to modify Section 230 that (a) satisfies this shaky coalition, (b) would be practical to comply with, and (c) would be constitutional.

Now, Section 230 critics are right: the law blurs the line between publisher and conduit. But this is not unique to Internet companies. The fact is, courts (and federal agencies) blurred the publisher-conduit dichotomy for fifty years for mass media distributors and common carriers as technology and social norms changed. Some cases that illustrate the phenomenon:

In Auvil v. CBS 60 Minutes, a 1991 federal district court decision, some Washington apple growers sued some local CBS affiliates for airing allegedly defamatory programming. The federal district court dismissed the case on the grounds that the affiliates are conduits of CBS programming. Critically, the court recognized that the CBS affiliates “had the power to” exercise editorial control over the broadcast and “in fact occasionally [did] censor programming . . . for one reason or another.” Still, case dismissed. The principle has been cited by other courts. Publishers can be conduits.

Conduits can also be publishers. In 1989, Congress passed a law requiring phone providers to restrict “dial-a-porn” services to minors. Dial-a-porn companies sued. In Information Providers Coalition v. FCC, the 9th Circuit Court of Appeals held that regulated common carriers are “free under the Constitution to terminate service” to providers of indecent content. The Court relied on its decision a few years earlier in Carlin Communications noting that when a common carrier phone company is connecting thousands of subscribers simultaneously to the same content, the “phone company resembles less a common carrier than it does a small radio station.”

Many Section 230 reformers believe Section 230 mangled the common law would like to see the restoration of the publisher-conduit dichotomy. As our research shows, that dichotomy had already been blurred for decades. Until advocates and lawmakers acknowledge these legal trends and plan accordingly, the reformers risk throwing out the baby with the bathwater.

Relevant research:
Brent Skorup & Jennifer Huddleston, The Erosion of Publisher Liability in American Law, Section 230, and the Future of Online Curation (Oklahoma Law Review).

Brent Skorup & Joe Kane, The FCC and Quasi–Common Carriage: A Case Study of Agency Survival (Minnesota Journal of Law, Science & Technology).

A series of recent studies have shown the centrality of social media bots to the spread of “low credibility” information online. Automated amplification, the process by which bots help share each other’s content, allows these algorithmic manipulators to spread false information across social media in seconds by increasing visibility. These findings, combined with the already rising public perception of social media as harmful to democracy, are likely to motivate some Congressional action regarding social media practices. In a divided Congress, one thing that seems to be drawing more bipartisan support is an antagonism to Big Tech.

Regulating social media to stop misinformation would mistake the symptoms of an illness for its cause. Bots spreading low quality content online is not a cause for declining social trust, but a result of it. Actions that explicitly restrict access to this type of information would likely result in the opposite of their intended effect; allowing people to believe more radical conspiracies and claim that the truth is censored.

A parallel for the prevalence of bots spreading information today is the high rates of media piracy that lasted from the late-1990s through the mid-2000s, but experienced a significant decline throughout this past decade (many of the claims by anti-piracy advocates of consistently rising US piracy fail to acknowledge the rise in file sizes of high quality downloads and the expansion of internet access, as a relative total of content consumption it was historically declining). Content piracy and automated amplification by bots share a relationship through their fulfillment of consumer demand. Just as nobody would pirate videos if there were not some added value over legal video access, bots would not be able to generate legitimate engagement solely by gaming algorithms. There exists a gap in the market to serve consumers the type of content that they desire in a convenient, easy-to-access form.

This fulfilment of market demand is what changed consumer interest in piracy, and it is what is needed to change interest in “low credibility” content. In the early days of the MP3 file format the music industry strongly resisted changing their business models, which led to the proliferation of file sharing sites like Napster. While lawsuits may have shut down individual file sharing sites, they did not alter the demand for pirated content, and piracy persisted. The music industry’s begrudging adoption of iTunes began to change these incentives, but pirated music streaming persisted. It was with legal streaming services like Spotify that piracy began to decline as consumers began to receive what they asked for from legitimate sources: convenience and cheap access to content. It is important to note that pirating in the early days was not convenient, malware and slow download speeds made it a cumbersome affair, but given the laggard nature of media industry incumbents, consumers sought it out nonetheless.

The type of content considered “low credibility” today, similarly, is not convenient, as clickbait and horrible formatting intentionally make such sites painful to use in order to maximize advertising dollars extracted. The fact that consumers still seek these sites out regardless is a testament to the failure of the news industry to cater to consumer demands.

To reduce the efficacy of bots in sharing content, innovation is needed in content production or distribution to ensure convenience, low cost, and subjective user trust. This innovation may come from the social media side through experimentation with subscription services less dependent on advertising revenue. It may come from news media, either through changes in how they cater content to consumers, or through changes in reporting styles to increase engagement. It may even come through a social transformation in how news is consumed. Some thinkers believe that we are entering a reputation age , which would shift the burden of trust from a publication to individual reporters who curate our content. These changes, however, would be hampered by some of the proposed means to curtail bots on social media.

The most prominent proposals to regulate social media regards applying traditional publisher standards to online platforms through the repeal of Section 230 of the Communications Decency Act, which in turn would make platforms liable for the content users post. While this would certainly incentivize more aggressive action against online bots – as well as a wide amount of borderline content – the compliance costs would be tremendous given the scale at which social media sites need to moderate content. This in turn would price out the innovators who would not be able to stomach the risks of having fewer bots than Twitter or Facebook, but still have some prevalent. Other proposals, such as the Californian ban on bots pretending to be human, reviving the Fairness Doctrine for online content, or antitrust action, range from unenforceable to counterproductive.

As iTunes, Spotify, Netflix, and other digital media platforms were innovating in the ways to deliver content to consumers, piracy enforcement gained strength to limit copyright violations, to little effect . While piracy as a problem may not have disappeared, it is clear that regulatory efforts to crack it down contributed little, since the demand for pirated content did not stem purely from the medium of its transmission. Bots do not proliferate because of social media, but because of declining social trust. Rebuilding that trust requires building the new, not constraining the old.

A few states have passed Internet regulations because the Trump FCC, citing a 20 year US policy of leaving the Internet “unfettered by Federal or State regulation,” decided to reverse the Obama FCC’s 2015 decision to regulate the Internet with telephone laws.

Those state laws regulating Internet traffic management practices–which supporters call “net neutrality”–are unlikely to survive lawsuits because the Internet and Internet services are clearly interstate communications and FCC authority dominates. (The California bill also likely violates federal law concerning E-Rate-funded Internet access.) 

However, litigation can take years. In the meantime ISP operators will find they face fewer regulatory headaches if they do exactly what net neutrality supporters believe the laws prohibit: block Internet content. Net neutrality laws in the US don’t apply to ISPs that “edit the Internet.”

The problem for net neutrality supporters is that Internet service providers, like cable TV providers, are protected by the First Amendment. In fact, Internet regulations with a nexus to content are subject to “strict scrutiny,” which typically means regulations are struck down. Even leading net neutrality proponents, like the ACLU and EFF, endorse the view that ISP curation is expressive activity protected by First Amendment.

As I’ve pointed out, these First Amendment concerns were raised during the 2016 litigation and compelled the Obama FCC to clarify that its 2015 “net neutrality” Order allows ISPs to block content. As a pro-net neutrality journalist recently wrote in TechCrunch about the 2015 rules,  

[A] tiny ISP in Texas called Alamo . . . wanted to offer a “family-friendly” edited subset of the internet to its customers. Funnily enough, this is permitted! And by publicly stating that it has no intention of providing access to “substantially all Internet endpoints,” Alamo would exempt itself from the net neutrality rules! Yes, you read that correctly — an ISP can opt out of the rules by changing its business model. They are . . . essentially voluntary.

The author wrote this to ridicule Judge Kavanaugh, but the joke is clearly not on Kavanuagh.

In fact, under the 2015 Order, filtered Internet service was less regulated than conventional Internet service. Note that the rules were “essentially voluntary”–ISPs could opt out of regulation by filtering content. The perverse incentive of this regulatory asymmetry, whereby the FCC would regulate conventional broadband heavily but not regulate filtered Internet at all, was cited by the Trump FCC as a reason to eliminate the 2015 rules. 

State net neutrality laws basically copy and paste from the 2015 FCC regulations and will have the same problem: Any ISP that forthrightly blocks content it doesn’t wish to transmit–like adult content–and edits the Internet is unregulated.

This looks bad for net neutrality proponents leading the charge, so they often respond that the Internet regulations cover the “functional equivalent” of conventional (heavily regulated) Internet access. Therefore, the story goes, regulators can stop an ISP from filtering because an edited Internet is the functional equivalent of an unedited Internet.

Curiously, the Obama FCC didn’t make this argument in court. The reason the Obama FCC didn’t endorse this “functional equivalent” response is obvious. Let’s play this out: An ISP markets and offers a discounted “clean Internet” package because it knows that many consumers would appreciate it. To bring the ISP back into the regulated category, regulators sue, drag the ISP operators into court, and tell judges that state law compels the operator to transmit adult content.

This argument would receive a chilly reception in court. More likely is that state regulators, in order to preserve some authority to regulate the Internet, will simply concede that filtered Internet drops out of regulation, like the Obama FCC did.

As one telecom scholar wrote in a Harvard Law publication years ago, “net neutrality” is dead in the US unless there’s a legal revolution in the courts.  Section 230 of the Telecom Act encourages ISPs to filter content and the First Amendment protects ISP curation of the Internet. State law can’t change that. The open Internet has been a net positive for society. However, state net neutrality laws may have the unintended effect of encouraging ISPs to filter. This is not news if you follow the debate closely, but rank-and-file net neutrality advocates have no idea. The top fear of leading net neutrality advocates is not ISP filtering, it’s the prospect that the Internet–the most powerful media distributor in history–will escape the regulatory state.

Lawmakers frequently hear impressive-sounding stats about net neutrality like “83% of voters support keeping FCC’s net neutrality rules.”   This 83% number (and similar “75% of Republicans support the rules”) is based on a survey from the Program for Public Consultation released in December 2017, right before the FCC voted to repeal the 2015 Internet regulations.

These numbers should be treated with skepticism. This survey generates these high approval numbers by asking about net neutrality “rules” found nowhere in the 2015 Open Internet Order. The released survey does not ask about the substance of the Order, like the Title II classification, government price controls online, or the FCC’s newly-created authority to approve of and disapprove of new Internet services.

Here’s how the survey frames the issue:

Under the current regulations, ISPs are required to:    provide customers access to all websites on the internet.    provide equal access to all websites without giving any websites faster or slower download speeds.  

The survey then essentially asks the participant if they favor these “regulations.” The nearly 400-page Order is long and complex and I’m guessing the survey creators lacked expertise in this area because this is a serious misinterpretation of the Order. This framing is how net neutrality advocates discuss the issue, but the Obama FCC’s interpretations of the 2015 Order look nothing like these survey questions. Exaggeration and misinformation is common when discussing net neutrality and unfortunately these pollsters contributed to it. (The Washington Post Fact Checker column recently assigned “Three Pinocchios” to similar net neutrality advocate claims.)

Let’s break down these rules ostensibly found in the 2015 Order.

“ISPs are required to provide customers access to all websites on the internet”

This is wrong. The Obama FCC was quite clear in the 2015 Order and during litigation that ISPs are free to filter the Internet and block websites. From the oral arguments:

FCC lawyer: “If [ISPs] want to curate the Internet…that would drop them out of the definition of Broadband Internet Access Service.” Judge Williams: “They have that option under the Order?” FCC lawyer: “Absolutely, your Honor. …If they filter the Internet and don’t provide access to all or substantially all endpoints, then…the rules don’t apply to them.”

As a result, the judges who upheld the Order said, “The Order…specifies that an ISP remains ‘free to offer ‘edited’ services’ without becoming subject to the rule’s requirements.”

Further, in the 1996 Telecom Act, Congress gave Internet access providers legal protection in order to encourage them to block lewd and “objectionable content.” Today, many ISPs offer family-friendly Internet access that blocks, say, pornographic and violent content. An FCC Order cannot and did not rewrite the Telecom Act and cannot require “access to all websites on the internet.”

“ISPs are required to provide equal access to all websites without giving any websites faster or slower download speeds”

Again, wrong. There is no “equal access to all websites” mandate (see above). Further, the 2015 Order allows ISPs to prioritize certain Internet traffic because preventing prioritization online would break Internet services.

This myth–that net neutrality rules require ISPs to be dumb pipes, treating all bits the same–has been circulated for years but is derided by networks experts. MIT computer scientist and early Internet developer David Clark colorfully dismissed this idea as “happy little bunny rabbit dreams.”  He pointed out that prioritization has been built into Internet protocols for years and “[t]he network is not neutral and never has been.” 

Other experts, such as tech entrepreneur and investor Mark Cuban and President Obama’s former chief technology officer Aneesh Chopra, have observed that the need for Internet “fast lanes” as Internet services grow more diverse. Further, the nature of interconnection agreements and content delivery networks mean that some websites pay for and receive better service than others.

This is not to say the Order is toothless. It authorizes government price controls and invents a vague “general conduct standard” that gives the agency broad authority to reject, favor, and restrict new Internet services. The survey, however, declined to ask members of the public about the substance of the 2015 rules and instead asked about support for net neutrality slogans that have only a tenuous relationship with the actual rules.

“Net neutrality” has always been about giving the FCC, the US media regulator, vast authority to regulate the Internet.  In doing so, the 2015 Order rejects the 20-year policy of the United States, codified in law, that the Internet and Internet services should be “unfettered by Federal or State regulation.” The US tech and telecom sector thrived before 2015 and the 2017 repeal of the 2015 rules will reinstate, fortunately, that light-touch regulatory regime.

The recently enacted Stop Enabling Sex Trafficking Act (SESTA) has many problems including that it doesn’t achieve its stated purpose of stopping sex trafficking . It contains a retroactivity clause that appears facially unconstitutional , but this provision would likely be severable by courts if used as the sole basis of a legal challenge. Perhaps more concerning are the potential First Amendment violations of the law.

These concerns go far beyond the rights of websites as speakers, but to the individual users’ content generation. Promoting sex trafficking is already a crime and a lawful restraint on speech. Websites, however, have acted broadly and quickly due to concerns of their new liability under the law and as a result lawful speech has also been stifled.

Given the controversial nature of the law it seems likely that a legal challenge is forthcoming. Here are three ideas about what a First Amendment challenge to the law might look like.

SESTA and Users’ Free Speech Rights

SESTA impacts individual users’ speech rights. As Elizabeth Nolan Brown writes, the law will create a chilling effect that could result in harming the very victims it claims to protect and could lead to further marginalizing minority viewpoints.

Despite their increasing presence and role in our everyday lives, Internet intermediaries, such as social media, are not public forums, but rather private actors. The recent Praeger case in California against YouTube has reinforced this point. As a result, they may choose to limit speech or actions in accord with terms of service or other policies.  Some would argue that moderation decision made in consideration of liability by these private actors do not constitute a violation of speech rights, but rather merely a modification of existing terms of service. However, this ignores both the chilling effects of such regulations and the fact that speech that would not be a violation of terms is likely to be removed as a result of broad interpretations of SESTA.

In the landmark case Reno v. ACLU , the Supreme Court recognized the problem of censoring online speech. In striking down the parts of the Communications Decency Act (CDA) other than Section 230’s liability protection, the Court stated, “[T]he CDA effectively suppresses a large amount of speech that adults have a constitutional right to receive and to address to one another. That burden on adult speech is unacceptable if less restrictive alternatives would be at least as effective in achieving the legitimate purpose that the statute was enacted to serve.” The results of SESTA have been a swift suppression of certain speech online and not just sex trafficking.

For example, Craigslist removed its entire personal section in response to the passage of SESTA. Ads that in no way could be considered a violation of either the terms of service or sex trafficking under federal laws were removed along with any potentially violative ads. Similarly, sex workers have expressed concerns sharing client information as a way to keep one another safe would be impossible under the statute as passed. Removing all this information also makes it more difficult for individuals trying  to help identify trafficking victims and facilitate their escape to find and assist victims and investigators. All of this information is lawful speech that will be either considered illegal or effectively eliminated by unnecessary burdens intermediaries must take to protect themselves from both criminal and civil liability.

The courts have generally favored allowing to disallowing speech. While minimal limits regarding time, manner, and place have been upheld in some cases and courts have found the state may regulate obscenity, speech restrictions are generally subject to strict scrutiny and must be narrowly tailored. SESTA uses broad definitions to classify what is considered sex trafficking and is likely to include both voluntary and involuntary interactions. Similarly the fact that the “participation in a venture” standard appears to set a low bar for an intermediary encourages an act first, question second behavior similar to that which has failed for the DMCA . To prevent liability under the statute, intermediaries must either increase moderation or cease moderating altogether. It is almost certain that lawful speech will regularly be caught up in such extreme moderation.

Finally, there are the concerns that chipping away at Section 230 liability opens the doors to broader Internet censorship . The Internet has been a stronghold of Free Speech where any idea can be expressed while well-intentioned laws like SESTA risk encouraging the idea that controversial or disliked speech can be censored.

Defining Intermediaries’ Editorial Control

Prior to Section 230 in Cubby v. Compuserve , the federal district court for the Southern District of New York found that Internet intermediaries act more like a distributor such as a bookstore or library than a traditional publisher. As a result, they have less control over the content created and distributed by their services than an editor or publisher would. Therefore, at common law, the intermediaries were found to have less liability for defamation or obscenity than a traditional publisher. This liability increases or decreases depending on the intermediary’s involvement with user generated content. Intermediaries who create or modify content are not acting as intermediaries and may be held liable if such content is illegal, such as sex-trafficking related content, even prior to SESTA.

The First Amendment Rights of Intermediaries

Intermediaries have free speech rights too. They may choose content to restrict or not restrict. Curation of content has been found to be protected as a form of speech for intermediaries such as search engines by several U.S. courts. In the pre-Internet Smith v. California case, the Supreme Court struck down the application of strict liability for obscene materials of a bookstore.  The court found that the lack of a knowledge requirement for criminal liability to attach was unconstitutional. SESTA requires knowledge but is vague regarding what knowledge an intermediary must have to be considered a participant in such a venture. Additionally, it gives broad power to state attorneys general to conduct investigation or take action with mere reasonable suspicion of a violation. One potential challenge would be whether the lack of a Good Samaritan clause and the vagueness regarding what constitutes knowledge in the statute violates the standards set in Smith .  Combined with the apparent protections of speech rights for intermediaries in the decisions to curate content, it may be possible for the intermediaries themselves to mount a First Amendment challenge.

Conclusion

SESTA has now become law, but it is almost certain it will face a constitutional challenge from users whose content was blocked or the intermediaries themselves on First Amendment grounds. In the past the courts have recognized the importance of maintaining free expression and a wide range of discourse online even when such content may be objectionable to many, one can only hope they would continue that line of thought if SESTA faces a First Amendment challenge.

Last Friday, law enforcement agencies shutdown Backpage.com. The website has become infamous for its role in sex trafficking, particularly related to underage victims, and its shutdown is rightly being applaud by many as a significant win for preventing sex trafficking online. This shutdown shows, however, that prosecutors had the tools necessary to go after bad actors prior to the passage of the Stop Enabling Sex Traffickers Act (SESTA) last month. Unfortunately, this is not the first time the government has pushed for regulation of technology knowing it already had the tools and information needed to build a case against bad actors.

The version of SESTA passed by Congress last month included a number of poorly thought through components including an ex post facto application and poorly articulated definitions, but it passed both houses of Congress with little opposition. In fact, because the law was seen as a must pass and linked to sex trafficking, the Senate even overwhelming rejected an amendment to provide additional funding for prosecuting such crimes. Even without being signed into law, SESTA has already resulted in Reddit and Craigslist removing communities from their platforms within days of its passage. What this most recent event shows is the government already had the tools to go after the bad actors like Backpage, but failed to use them as Congress debated and passed a law that chipped away at the protection for the rest of the Internet and gave the government even broader powers.

This is not the first time that the government has encouraged through either its action or inaction damaging regulation of disruptive technology while knowing that it had tools at its disposal that could achieve the desired results without the need for an additional regulatory burden. In 2016, the government argued following the San Bernadino shootings that it need more access to encrypted devices like the iPhone when Apple refused to comply with a writ compelling it to unlock the shooters’ phones. The Senate responded to the controversy by proposing a bill that would require business like Apple to assist authorities in gaining access to encrypted devices. Thankfully, because the FBI was able to gain the information needed without Apple through a third party vendor, such calls largely diminished and the legislation never went anywhere.  Now, a recent Office of the Inspector General report has revealed the FBI “testified inaccurately or made false statements” regarding its ability to gain data from the encrypted iPhone.

It is highly concerning that when the government has the tools needed to engage in action to stop bad actors, but desires more regulatory power over tech it chooses to pursue regulation for all instead of using the proper tools it already has to pursue the bad actors. Rather than gaining new tools that risk ruining innovation, the government should first exhaust the tools they have to prosecute those bad actors. When they do use these tools against the likes of MyRedbook, Rentboy, and now Backpage, the prosecutions have been by and large successful. This continued pattern of behavior should raise heightened concerns about  calls for greater regulation of technology and whether the trade-offs such regulation would require are needed.

Neither the government’s desire for more regulation nor its negative impact is limited to technology. Research from the Mercatus Center has shown that the cumulative effect of regulation has slowed GDP growth by 0.8% per year since 1980. Particularly for new startups these regulatory burdens increase the cost of even entering the increasingly global marketplace due to both increased compliance costs and fears of company ending litigation.

With the awareness that these additional regulations are often unnecessary and harmful to both technology and the economy more generally, there should be heightened concern for calls to give regulators additional tools in light of specific events. These calls for regulation have once again arisen with recent fatal Uber autonomous vehicle accident and the Facebook scandals. These regulations may actually make problems worse not better by creating a regulated monopoly that prevents new entrants from improving quality and increasing competition. As Mark Zuckerberg noted while answering a question during Congressional testimony yesterday when there are more rules it is easier for larger companies to comply with them than smaller companies.  Additional regulations make it more difficult for us to get the next Facebook, the next Google, or the next Uber.

The overall framework of Permissionless Innovation put in effect during the Clinton administration has allowed the Internet to flourish and the US to become a global leader in Internet innovation and we must not let the failure to use the tools available deceive us into believing that such an environment does not work. Regulation is sometimes necessary, but over regulation, particularly of technology, poses significant risks that must be considered in more than just a reactive fashion.

SESTA passed the Senate last week after having previously passed the House. President Trump is expected to sign it into law despite the opposition to this version of the bill from the Department of Justice. As I have previously written about, there are a great deal of concerns about how the bill may actually make it harder to address online sex trafficking and more generally impact innovation on the Internet.

The reality is that we are looking at a post-SESTA world without the full protection of Section 230 and that reality will likely end up far from the best case scenario, but hopefully not fully at the worst. Intermediaries, however, do not have the luxury to wait around and see how the law actually plays out, especially given its retroactive provision. As a result, Reddit has already deleted a variety of sub-reddits and Craigslist has closed its entire personals section. One can only imagine the difficult decisions facing the creators of dating apps or messaging services.

So what can we expect to happen now…

1.    Questions remain about how often the law will be used, and when civil cases will be brought

Prosecutors just a few years ago were given additional criminal resources to prosecute sex trafficking in the SAVE Act. Yet, these tools have rarely been used due to the difficulty in prosecuting such crimes. Similarly, most civil litigation settles out of court. Especially given the potential PR nightmares of being seen as not believing victims or favoring bad actors if a civil case does go to trial, there will be a great deal of pressure on intermediaries to settle out of court whether they engaged in unlawful actions or not. The push for settlement will likely be even stronger for smaller companies who lack the resources to hire legal teams, fund litigation, and risk greater damage to the business.

2.   However, at some point SESTA will likely end up in court and likely face a constitutional challenge.

The response on the part of websites to the requested changes seems to have been swift and far-reaching. Given that SESTA presents First Amendment challenges and has a most likely unconstitutional retroactive provision, the question seems to be who and when the law will be challenged in courts.

The retroactive nature of the law appears facially unconstitutional. It is, however, likely the courts would be able to sever this provision from the rest of the law. This would fix some of the minor issues with establishing liability after a decision regarding moderation was made, but would not fix the broader innovation and speech quashing concerns of the law.

The First Amendment challenges could come either from sex workers whose lawful speech is being silenced or from those not at all related to sex work whose innocent actions were censored as a result of an intermediary’s low risk tolerance due to increased liability under SESTA.

3.    Big intermediaries like Facebook and Google will adjust, but new intermediaries may struggle to get off the ground.

Facebook deletes over 1 million accounts a day. Various tech and app companies are estimated to employ over 100,000 moderators to evaluate user generated content. This work is deeply disturbing and has a high human toll for those engaged in it as other technology has not been able to replace the ability of human moderators to make certain distinctions. Large companies might be able to adapt by hiring more moderators or deleting certain user communities for potential liability raising areas, but smaller companies will be even less able to compete and adapt.

SESTA may prevent us from getting the next Google, Facebook, or Paypal for three key reasons. First, it raises the initial cost of launching a product that has user generated content by requiring additional moderators just to get off the ground. Second, it is likely to make funders less likely to invest in new intermediaries like messaging and dating services if they are concerned that the company is likely to get sued. Third, it may prevent existing small and mid-size tech or app companies in areas like social media or messaging from expanding or innovating in areas that are likely to have interactions between users due to concerns about liability.

For all the concerns that tech is getting too centralized in a few companies, there seems to be little attention paid to the fact that raising the liability risks through laws such as SESTA may result in a scenario where only those few big companies can comply.

4.    It sets an uneasy precedent for further eroding Section 230.

This is perhaps the greatest concern. Sex-trafficking is evil, but prosecutors had the tools to go after it and Section 230 already had a carve out for federal crimes. SESTA signals that a legislative reaction to a single or few bad actors’ actions online can result in chipping away at the protection that has allowed the Internet to flourish. It shows that such actions are often not narrowly tailored due to their reactive nature. Especially as there are growing concerns about various individual actors, we must remember that broad legislation risks making it difficult for good actors and new challengers to try to take their place. A post-SESTA world signals that while Section 230 may still exist, it is far to easily eroded for all when concerns about the bad actions of a few arise.

What happens over the next few months and years as both new and existing intermediaries try to adapt will greatly influence the future of the Internet and its ability to be a tool for global connectedness. As Senator Wyden said following the rejection of an amendment to SESTA to fund sex trafficking prosecutions, “I anticipate having to turn back to this topic in short order after the effects of this bill become clear.” How swiftly those effects are felt by everyone and whether the reality of their damage to innovation is clear to policymakers remains unknown, but that such effects will occur in one form or another cannot be disputed.

While the Net Neutrality debate has been in the foreground, Congress has been quietly moving forward legislation that risks fundamentally modifying the liability protection for Internet intermediaries like Facebook, Google, and PayPal, and forever changing the Internet. The proposed legislation has good intentions of stopping sex trafficking, but in an effort to stop a few bad actors the current overly broad version of the bill risks not only stopping the next Internet innovation, but also failing to achieve even this laudable goal.

Where Are We Now: A Legislative Update

As I have written earlier, the House and the Senate version each introduced bills nobly aimed at preventing and fighting sex trafficking. The House bill, FOSTA, was amended during the committee process and these significant changes minimized many of the most concerning elements of the original version of the legislation. The bill still had many flaws including standards that remained vague and did not account for a website’s size, but it was generally applauded as a significant step towards achieving its goal while minimizing the damage to free expression on the Internet. The Senate bill, SESTA, retained many of the concerns of the initial FOSTA bill. Before the House voted, FOSTA was amended to include all elements of SESTA both good and bad. The bill with SESTA attached passed the House and now proceeds to the Senate where a vote is expected next week.

The Continuing Problems of FOSTA/SESTA

According to Internet law professor Eric Goldman, unfortunately the House passed FOSTA now represents the worst of both worlds and could have far reaching implications not just for those engaged in detestable practices but also for advocates, social media, and free speech online more generally. The current version of the bill has also been criticized by many including not only the tech community, but also the prosecutors at the Department of Justice.

There are at least three primary issues remaining in the FOSTA/SESTA legislation as proposed.

First, it could make the problem of identifying and rescuing victims more difficult for advocates. This is for two main reasons. As law professor Ariel Levy pointed out even if the bill succeeds in removing sex trafficking online, it will only push the true perpetrators of these acts further underground making it harder for those seeking to monitor and prosecute such crimes to find victims. It also risks silencing the spread of information to help victims due to broad language in the law and the difficulty companies would have in distinguishing such messages. Finally, the law does not distinguish forced from voluntary transactions. Advocates for sex workers have expressed concerns that the law would prevent the sharing of information that has increased safety.

Second, it could actually make it more difficult for prosecutors to go after perpetrators of these crimes. The Department of Justice letter points out that the vague language such as “participation in a venture” will make it harder to prosecute wrongdoers. As I have previously discussed, prosecutors have the tools and should be encouraged to use them. Mike Masnick recently pointed out that while the bill creates a new crime, it is already illegal to engage in and advertise sex-trafficking. The current vagueness and imposition of new liability on third parties not actively engaged in trafficking could make it harder for prosecutors to use the tools they have to go after the actual traffickers.

Finally, as Rep. Justin Amash questioned in the immediate aftermath of its passage the bill as currently written could easily be interpreted as allowing for ex post facto liability and prosecutions. The version passed by the House expressly allows the prosecution of actions that would have been illegal under the law even if the actions occurred years before its passage. If such provisions were enforced, it’s plausible the courts could find the statute facially unconstitutional.

Potential Solutions

Section 230 immunity has allowed the Internet to flourish for over 20 years. Without such protections, it is unlikely that many user generated communities like social media sites or messaging services would have developed.  Since the Senate has not yet voted on the bill, there is still time to leave Section 230 as it currently functions or for amendments that could minimize the risks described above.

First, as suggested by the Department of Justice letter attention should be given to vague definition of participation to limit the application of the law to only those who actively engage in such acts. The current language means that a search engine, payment processor, or social media site could be found liable for even a single transaction by a user. Clear definitions are particularly important given they impact not only civil liability but also the creation of a new crime.

Second, the intent requirements could be raised to limit the law only to those with truly bad intentions and protect Good Samaritan actors who accidentally make a mistake. The current version has a relatively low requirement for liability. A recent Wall Street Journal editorial pointed out that an attorney would only need to show that the website “should have known” not that they actually knew this behavior was going on in order to bring a lawsuit. As a result, intermediaries are most likely to engage in aggressive censorship. This could result in wrongfully silencing advocates as discussed above. Of course, others could choose not to engage in moderating at all out of a fear that they will be found to have knowledge. Ideally, a provision to protect moderator actions and a heightened mens rea requirement would minimize these risks.

Third, remove any ex post facto applications of the statute. A website could not have taken additional steps to comply with a law that existed prior to its passage, so should only reasonably be held liable for actions that occur since the law’s passage. Even for seemingly innocuous social media websites like Facebook or search engines like Google the new standard would require significantly more resources devoted to monitoring than they already engage in. Given that the law would undo two decades of status quo for moderation, it seems providing intermediaries a few months to insure they have the necessary resources is a reasonable change.

Section 230 has worked to allow the Internet to flourish in ways that could not have been predicted 20 years ago. Any changes to Section 230 liability protection are likely to have far reaching implications for the Internet and innovation. While these changes may be brought with good intentions, they risk fundamentally changing  nature of new communications tools and doing quite a bit more than just targeting bad actors.

Internet regulation advocates lost their fight at the FCC, which voted in December 2017 to rescind the 2015 Open Internet Order. Regulation advocates have now taken their “net neutrality” regulations to the states.

Some state officials–via procurement contracts, executive order, or legislation–are attempting to monitor and regulate traffic management techniques and Internet service provider business models in the name of net neutrality. No one, apparently, told these officials that government-mandated net neutrality principles are dead in the US.

As the litigation over the 2015 rules showed, o ur national laissez faire policy towards the Internet and our First Amendment guts any attempt to enforce net neutrality.  Recall that the 1996 amendments to the Communications Act announce a clear national policy about the Internet:

It is the policy of the United States . . . to preserve the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.

In fact, that 1996 law was passed in order to encourage ISPs to filter objectionable content.

Further, regulators cannot prevent ISPs from exercising their First Amendment rights to curate the Internet. As Prof. Stuart Minor Benjamin wrote for the Harvard Law Review Forum in 2014,

If we really want to prevent Internet access providers from being speakers, we are going to have to radically reshape the Supreme Court’s First Amendment jurisprudence and understandings.

No radical reshaping of the First Amendment has occurred. For all these reasons, the Obama FCC attorney was forced to concede that

If they [that is, ISPs] filter the Internet . . . the [2015 Open Internet] rules don’t apply to them. 

Even Title II supporters EFF and the ACLU acknowledge in their FCC joint filing that ISPs are speakers who can filter content and escape Title II regulation.

ACLU & EFF throw in the towel—ISPs may block content under the 1st Amend. & are exempt from Title II. #netneutrality https://t.co/xU7uWfPIn4 pic.twitter.com/ZrXjs8bUtS

— Brent Skorup (@bskorup) September 1, 2017

At the end of the day, net neutrality, having lost its original definition, is simply a re-branding of Internet regulation.

State Internet regulations, therefore, are at odds with federal law and policy.  Let’s set aside federal preemption for the moment (Seth Cooper explained why preemption likely kills most of these state Internet regulations). There are other arguments for why states can’t impose baby “net neutrality” bills. 

Net neutrality bills likely violate the law

The state “net neutrality” bills and executive orders represent common carriage regulation. State officials make no attempt to hide this since they largely copy-and-paste the nondiscrimination obligations directly from the 2015 Open Internet Order. Here’s the problem for states: regulators can’t impose common carrier obligations on non-common carriers.

When nondiscrimination principles deprive operators of control of content, that amounts to common carriage. This was established in a 1979 Supreme Court case, Midwest Video II. In that case, the Supreme Court struck down common carriage obligations on cable operators, who are non-common carriers. The Court said ,

With its access rules, however, the Commission has transferred control of the content of access cable channels from cable operators to members of the public who wish to communicate by the cable medium. …The access rules plainly impose common-carrier obligations on cable operators.

The FCC, the Court said, had no authority to transform them into common carriers.

In fact, this is why the 2010 Open Internet Order was struck down in Verizon v. FCC. There, relying on Midwest Video II, the DC Circuit held that the net neutrality principles couldn’t be enforced on non-common carriers. As the DC Circuit said of the FCC’s common carrier obligations for ISPs: “Midwest Video II is indistinguishable.”

State “net neutrality” regulations will likely fail for the same reason.  The 2015 rules were upheld because “broadband Internet access service” was classified as a Title II common carrier service. “Broadband Internet access service” providers will no longer be common carriers once the 2017 Restoring Internet Freedom Order takes effect. By imposing common carrier rules on non-common carriers, states run afoul of Midwest Video II and Verizon.

Net neutrality bills balkanize the Internet

State-based Internet regulation is also bad policy, and many who support net neutrality principles–like Google–oppose this legal regime. Internet regulation advocates, by encouraging regulation state-by-state and city-by-city, have finally dispensed with the fiction that “net neutrality” is about the “open Internet.” In their eagerness to have someone, anyone regulate the Internet, these advocates are willing to balkanize the US Internet into dozens, or even hundreds, of splinternets, each with a different local or state regulator.

The Montana governor, for instance, encouraged every state and city to regulate the Internet, even providing a customizable template:

Any city or state can do this. We made you a template: https://t.co/yYgQTWdat1 #NetNeutrality https://t.co/dz3BFiN0uN

— Steve Bullock (@GovernorBullock) January 22, 2018

Further, net neutrality rules are not easy to apply and interpret, particularly the “catch-all” Internet conduct standard. Net neutrality supporters take vastly different stances on identical ISP conduct.

One illustration: the common practice of zero rating by mobile providers. One prominent net neutrality supporter (then-FCC chairman Tom Wheeler) said T-Mobile’s zero rating was “highly innovative and highly competitive.” Another (Prof. Susan Crawford) said it is “anti-competitive,” “dangerous,” and “malignant” and should be ended immediately. There were many advocates in both camps and everywhere in between.

Given the wide divergence of views on a single issue, dozens of “net neutrality” laws would create innumerable contradictions about what is allowed and disallowed online. The fragmented Internet and legal uncertainty would be particularly damaging to small app companies and competitive ISPs, who don’t have hallways of lawyers to ensure compliance, and who use or plan to use traffic priority techniques for gaming, disability services, VoIP, and driverless cars.

For the global, stateless Internet, having state and city CIOs create their own custom Internet regulation interpretations would destroy what made the Internet transformative–a permissionless, global network free of legacy regulations. State legislatures and governors, by ramming through “net neutrality,” are committing to waste countless taxpayer dollars in battling the federal government and telecom companies in (probably unwinnable) litigation. Their “best-case” scenario: a few states win in court and splinter the Internet.

Hopefully cooler heads will prevail and put state energies and treasure into doing something constructive about broadband, like urging reform of the $8.8 billion universal service fund or improving permitting processes and competition.

The house version of the Stop Enabling Sex Trafficking Act (SESTA), called the Allow States and Victims to Fight Online Sex Trafficking Act (FOSTA), has undergone significant changes that appear to enable it to both truly address the scourge of online sex trafficking and maintain important internet liability protection that encourages a free and open internet. On Tuesday, this amended version passed the House Judiciary Committee. Like most legislation, this latest draft isn’t perfect. But it has made significant steps towards maintaining freedom online while addressing the misdeeds of a few.

The Good

First, the new version creates a new crime that targets online sex traffickers and those wrong-doers who intentionally promote or facilitate their actions. Earlier versions of the House and Senate sex trafficking bills created mens rea, or state of mind, issues whereby a website was compelled to engage in strict moderation for fear of something “falling through the cracks” while encouraging good behavior on the part of intermediaries. The new FOSTA proposal substitutes a higher standard, which largely obviates these concerns.

The revised bill also clearly focuses on sex trafficking and online prostitution rather than attacking potential “bad actions” online more generally. Even so, some are concerned about the impact this revised focus may have on consensual transactions or protected (even if objectionable) speech. However, combined with the creation of a new crime under the Mann Act, it appears to remove most of the early concerns that the new law could be applied too broadly and chip away at Section 230. Indeed, the language of the new bill makes it clear that Section 230 was “never intended to provide legal protection to websites that unlawfully promote and facilitate prostitution and contribute to sex trafficking.”

The revised bill creates civil liability only when a violation of the new criminal law has already occurred. This prevents someone from going after an intermediary merely because they have “deeper pockets” than the actual perpetrators. By requiring an intermediary also be guilty of a criminal violation, it limits the likelihood that individuals would be successful in such suits except in cases where the website had knowingly facilitated or actively encouraged such violations of the law.

Finally, the revised FOSTA relies on a national standard instead of a patchwork of state law claims. Given the truly global nature of the Internet, this provides greater certainty for intermediaries regarding under what standard they will be held liable.

The Remaining Questions/Concerns

The current version of the bill uses a standard of 5+ victims for the new criminal enhancement. But there is a problem with using a raw number of victims, as Eric Goldman points out.  He lays out a thought experiment: let’s say that a  larger website like Google or Facebook, has  0.01% of its usage dedicated to prostitution. That’s about 100,000 people. Goldman points out that even if these companies were 99.99% compliant in taking down this activity—a worthy feat, to be sure—some would surely still fall through the cracks. The 5+ standard could make the social platforms look like “hotbeds of prostitution activity” despite their best intentions. A simple solution would be to switch from a raw number of “victims” to a percentage of users or revenues before attaching criminal or civil liability.

Additionally, there are some concerns about whether the new law could still make things worse for victims. As one advocate wrote, putting victims on the street rather than online may make them much more likely to be subject to violence and may make it more difficult to identify and assist trafficking victims. Unfortunately, the dangers and harms associated with trafficking and sex work cannot be resolved by a single bill.

Like most legislation, FOSTA is not perfect, but the current version does avoid the most damaging elements of earlier iterations. The changes also show that legislators are becoming aware of the possible unintended consequences that broader legislation could lead to.

As I have previously written about, a bill currently up for debate in Congress runs the risk of gutting critical liability protections for internet intermediaries. Earlier today the Stop Enabling Sex Traffickers Act passed out of committee with an amendment attempted to remedy some of the most damaging changes to Section 230 in the original act. While this amendment has gained support from some industry groups, it does not fully address the concerns regarding changes to intermediary liability under Section 230. While the amended version shows increased awareness of the far reaching consequences of the act, it does not fully address issues that could have a chilling effect on speech on the internet and risk stifling future internet innovation.

• Good Samaritan Provision

As Eric Goldman points out, the amended version expressly retains part the Good Samaritan provisions of Section 230 for removals, but it still enables new liability for user publication. As a result, the new amendment only partially preserves Good Samaritan mechanisms and does not fully address concerns about good faith attempts to avoid the new liability.

• Knowledge Standard

The amended version that cleared committee clarifies the knowledge standard of the earlier bill by stating that for liability to attach the intermediary the intermediary must have participated by “knowingly assisting, supporting, or facilitating a violation.” This improves but does not fully mitigate the damage that establishing new liability could do to internet free speech. As EFF writes, facilitates legally means “to make easier or less difficult” and would include a huge swath of innocuous products, websites, and activities.

This standard is particularly dangerous for online dating and messaging services. For example, if a traffickers used a messaging service to communicate this could be seen as facilitating because it made it easier to communicate. Dating services which set up meetings could also be seen as facilitators if a bad actor used their service to conduct human trafficking. As Mike Masnick at TechDirt argues, there is little certainty in the amended version of what “knowingly” means and it may be as low a standard as general knowledge or media reports that your website was at some point used (or allegedly used) by sex traffickers.

• The Retroactivity Provision

The new version does not clear up concerns about retroactivity. In both the amended and original versions, the bill states that it applies “regardless of whether the conduct alleged occurred, or is alleged to have occurred, before, on or after such date of enactment.”  As a result companies are open to civil and criminal liability for conduct that did not have such liability when it occurred.

While the amended SESTA signals a recognition that the bill needs to more narrowly tailored,  it leaves internet intermediaries with the same two choices if enacted.

The first option for intermediaries would be to engage in an aggressive takedown process like they do for copyright claims under the DMCA. This reaction is to take down questioned content first and ask whether it should have been taken down later.  As Masnick notes, however, the DMCA has much clearer provisions for when content must be taken, but still there are seemingly rampant issues with false claims. Such a situation would only be worse under SESTA especially for social media, search engines, and dating websites. Some websites might choose to quit operating rather than engage in the high level of moderation that would be necessary. Because the bill applies to all sizes of companies without any limitations for the amount staff or users, this is more likely to have a negative impact on smaller or more innovative services that might one day become the next Facebook or Google. These companies do not have the same manpower to engage in an aggressive monitoring for their user base and as a result may have more difficulty entering the market if there are greater compliance burdens. Those that did continue would only be those who could afford to devote large number of staff and legal resources to monitoring and determining the accuracy of claims.

The second option is to avoid the cooperation and self-monitoring that websites engage in now.  In a recent interview regarding Russian election ads, Senate Majority leader Mitch McConnell stated that tech should be “more interested in cooperating with law enforcement.” SESTA, however, provides the opposite incentive. Cooperation in investigations would show knowledge and open the intermediary up to further civil liability. As a result, intermediaries might be discouraged from future cooperation.

The amended version will now head to the Senate floor for debate. The bill has a noble goal of making sex trafficking more difficult and this revised version shows progress towards protecting intermediary liability. Still, it addresses the issue more broadly than needed and risks fundamentally changing the internet.

[originally published on Plaintext on June 21, 2017.]

This summer, we celebrate the 20th anniversary of two developments that gave us the modern Internet as we know it. One was a court case that guaranteed online speech would flow freely, without government prior restraints or censorship threats. The other was an official White House framework for digital markets that ensured the free movement of goods and services online.

The result of these two vital policy decisions was an unprecedented explosion of speech freedoms and commercial opportunities that we continue to enjoy the benefits of twenty years later.

While it is easy to take all this for granted today, it is worth remembering that, in the long arc of human history, no technology or medium has more rapidly expanded the range of human liberties — both speech and commercial liberties — than the Internet and digital technologies. But things could have turned out much differently if not for the crucially important policy choices the United States made for the Internet two decades ago.

First, on June 26, 1997, the Supreme Court handed down its landmark decision in Reno v. ACLU, which struck down the Communications Decency Act’s provisions seeking to regulate online content under the old broadcast media standard. The Court concluded that there was “no basis for qualifying the level of First Amendment scrutiny that should be applied to this medium” and rejected the congressional effort to pigeonhole this exciting new medium into the archaic censorship regimes of the past.

The Reno decision was tremendously important in protecting online speakers from the chilling effect of government “indecency” regulations. The decision also set a strong legal precedent and was cited in countless subsequent decisions involving not only online speech, but also efforts to regulate video game content.

Second, in July 1997, the Clinton Administration released The Framework for Global Electronic Commerce, a document that outlined the US government’s new policy approach toward the Internet and the emerging digital economy. The Framework was a bold vision statement that endorsed comprehensive online freedom of exchange, saying that “the private sector should lead [and] the Internet should develop as a market driven arena not a regulated industry.” The Administration rejected a restrictive regulatory regime for commercial activities and instead recommended reliance on civil society, contractual negotiations, voluntary agreements, and industry self-regulation.

To “avoid undue restrictions on electronic commerce,” the vision statement recommended that “parties should be able to enter into legitimate agreements to buy and sell products and services across the Internet with minimal government involvement or intervention.” But, “[w]here governmental involvement is needed, its aim should be to support and enforce a predictable, minimalist, consistent and simple legal environment for commerce.”

Taken together, the Reno decision and the Clinton Administration’s Framework acted as a Magna Carta moment for the Internet and digital technologies. It signaled that “permissionless innovation” would become America’s governance stance toward online speech and commerce.

As I defined it in a book on the subject, permissionless innovation, “refers to the notion that experimentation with new technologies and business models should generally be permitted by default. Unless a compelling case can be made that a new invention will bring serious harm to society, innovation should be allowed to continue unabated and problems, if any develop, can be addressed later.” The primary advantage of permissionless innovation as a governance disposition is that it sends a clear green light to citizens telling them they are at liberty to pursue their own interests and passions, free from the suffocating grip of prior restraints on free speech and free exchange.

But the Reno decision and the Clinton Administration’s Framework are not the only critical policy decisions that helped enshrine permissionless innovation as the lodestar of online policy in the US. In the mid-1990s, the Clinton Administration made the decision to allow open commercialization of the Internet, which was previously just the domain of government agencies and university researchers. Even more crucially, when Congress passed and President Bill Clinton signed into law the Telecommunications Act of 1996, lawmakers made it clear that traditional analog-era communications and media regulatory regimes would generally not be applied to the Internet.

The Telecom Act also included an obscure provision known as “Section 230,” which immunized online intermediaries from onerous liability for the content and communications that traveled over their networks. Section 230 was hugely important in that it let online speech and commerce flourish without the constant threat of frivolous lawsuits looming overhead. Internet scholar David Post has argued that “it is impossible to imagine what the Internet ecosystem would look like today without [Section 230]. Virtually every successful online venture that emerged after 1996 — including all the usual suspects, viz. Google, Facebook, Tumblr, Twitter, Reddit, Craigslist, YouTube, Instagram, eBay, Amazon — relies in large part (or entirely) on content provided by their users, who number in the hundreds of millions, or billions,” he notes. It is unlikely that the vibrant marketplace of online speech and commerce we enjoy today could have existed without the protections afforded by Section 230.

Finally, in 1998, another important legislative development occurred when Congress passed the Internet Tax Freedom Act, which blocked all levels of government in the US from imposing discriminatory taxes on the Internet. That made it clear that the Net would not be milked as a “cash cow” the way previous communications systems had been.

So, let’s recap how policymakers generally got policy right for the Internet in the mid-1990s by enshrining permissionless innovation as the law of the land:

• The Executive Branch set the tone for online freedom by fully privatizing the underlying network and then establishing a governance vision based upon minimal government interference with online speech and exchange.
• The Legislative Branch generally endorsed the Clinton Administration’s vision for the Internet and digital technologies by ensuring that new policies would not be based upon the failed regulatory and tax policies of the past.
• The Judicial Branch upheld the centrality of the First Amendment in the Information Age and made it clear that this new medium for speech would be granted the strongest protection against government encroachments on freedom of speech and expression.

The combined effect of these wise, bipartisan policy decisions was that the Net and digital tech were “born free” instead of being born into regulatory captivity. We continue to enjoy the fruits of these freedoms today as citizens here in the US and across the world take advantage of the unprecedented ability to connect and communicate to pursue their passions and interests as they see fit.

There’s still more work to be done, however. Online platforms and digital technologies continue to come under attack from regulatory activists both here and abroad. Many governments continue to push back against these online speech and commercial freedoms, meaning we’ll need to redouble our efforts to highlight and defend the benefits of preserving these important victories.

Finally, as the underlying drivers of the Digital Revolution continue to spread into other segments of the economy, these freedoms will come into conflict with older top-down regulatory regimes for automobiles, aviation, medical technology, finance, and much more. This will create an epic conflict of governance visions between the Internet’s permissionless innovation model versus the precautionary, command-and-control regulatory regimes of the industrial age. We already see tension at work in policy deliberations over the Internet of Things, “big data,” driverless cars, commercial drones, robotics, artificial intelligence, 3D printing, virtual reality, the sharing economy, and others.

If policymakers hope to preserve and extend the benefits of the hard-fought victories of the Internet’s past twenty years, they will need to restate and reinvigorate their commitment to permissionless innovation to help spur the next great technological revolutions in these and other fields.

If Congress and the President wanted to prevent intrusive regulation of the Internet, how would they do it? They know that silence on the issue wouldn’t protect Internet services. As Congress learned in the 1960s and 1970s with cable TV, congressional silence, to the FCC, looks like permission to enact a far-reaching regulatory regime.

In the 1990s, Congress knew the FCC would be tempted to regulate the Internet and Internet services and that silence would be seen as an invitation to regulate the Internet. Congress and President Clinton therefore passed a 1996 law, Section 230 of the Communications Decency Act, which stated:

It is the policy of the United States… to preserve the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.

But this statement raised the possibility that the FCC would regulate Internet access providers and would claim (as FCC defenders do today) they were not regulating “the Internet,” only access providers. To preempt such sophistry, Congress added that the “interactive computer services” shielded from regulation include:

specifically a service or system that provides access to the Internet….

Congress proved prescient. For over a decade, as the FCC’s traditional areas of regulation waned in importance, advocates and FCC officials have sought to regulate Internet access providers and the Internet. After two failed attempts to regulate providers and enforce net neutrality norms, the FCC decided to regulate Internet access providers with Title II, the same provisions regulating telephone and telegraph providers. Section 230 featured prominently in the dissents of commissioners Pai and O’Rielly who both noted that the Open Internet Order was a simple rejection of the plain words of Congress. Nevertheless, two judges on DC Circuit Court of Appeals blessed those regulations and the Open Internet Order in 2016.

If “unfettered from Federal regulation” means anything, doesn’t it mean that the FCC cannot use Title II, its most stringent regulatory regime, to regulate Internet access providers?  Is there any combination of words Congress could draft that would protect Internet access providers and Internet services from Title II?

There is a pending appeal challenging the Open Internet Order before the DC Circuit and after that is appeal to the Supreme Court. The Supreme Court, in particular, might be receptive to a common-sense argument that “unfettered from Federal regulation” is hazy around the edges but it cannot mean regulation of ISPs’ content, services, protocols, network topology, and business models.

I understand the sentiment that a net neutrality compromise is urgently needed to save the Internet from Title II. But until the Open Internet Order appeals have concluded, I think it’s premature to compromise and grant the FCC permanent authority to regulate the Internet with vague standards (e.g., no one knows what “reasonable throttling” means). A successful appeal could mean a third and final court loss for net neutrality purists, thereby restoring Section 230’s free-market protections for the Internet. Until the Supreme Court denies cert or agrees with the FCC that up is down, black is white, and agencies can ignore clear statutes, I’m not persuaded that Congress should nullify its own deregulatory language of Section 230 with a net neutrality compromise.

Title II allows the FCC to determine what content and media Internet access providers must transmit on their own private networks, so the First Amendment has constantly dogged the FCC’s “net neutrality” proceedings. If the Supreme Court agrees to take up an appeal from the DC Circuit Court of Appeals, which rejected a First Amendment challenge this summer, it will likely be because of Title II’s First Amendment deficiencies.

Title II has always been about handicapping ISPs qua speakers and preventing ISPs from offering curated Internet content. As former FCC commissioner Copps said, absent the Title II rules, “a big cable company could block access to an investigative report about its less-than-stellar customer service.” Tim Wu told members of Congress that net neutrality was intended to prevent ISPs from favoring, say, particular news sources or sports teams.

But just as a cable company chooses to offer some channels and not others, and a search engine chooses to promote some pages and not others, choosing to offer a curated Internet to, say, children, religious families, or sports fans involves editorial decisions. As communications scholar Stuart Benjamin said about Title II’s problem, under current precedent, ISPs “can say they want to engage in substantive editing, and that’s enough for First Amendment purposes.”

Title II – Bringing Broadcast Regulation to the Internet

Title II regulation of the Internet is frequently compared to the Fairness Doctrine, which activists used for decades to drive conservatives out of broadcast radio and TV. As a pro-net neutrality media professor explained in The Atlantic last year, the motivation for the Fairness Doctrine and Title II Internet regulation is the same: to “rescue a potentially democratic medium from commercial capture.” This is why there is almost perfect overlap between the organizations and advocates who support the Fairness Doctrine and those who lobbied for Title II regulation of the Internet.

These advocates know that FCC regulation of media has proceeded in similar ways for decades. Apply the expansive “gatekeeper” label to a media distributor and then the FCC will regulate distributor operations, including the content transmitted. Today, all electronic media distributors–broadcast TV and radio, satellite TV and radio, cable TV, and ISPs–whether serving 100 customers or 100 million customers, are considered “gatekeepers” and their services and content are subject to FCC intervention.

With broadband convergence, however, the FCC risked losing the ability to regulate mass media. Title II gives the FCC direct and indirect authority to shape Internet media like it shapes broadcast media. In fact, Chairman Wheeler called the Title II rules “must carry–updated for the 21st century.”

The comparison is apt and suggests why the FCC can’t escape the First Amendment challenges to Title II. Must-carry rules require cable TV companies to transmit all local broadcast stations to their cable TV subscribers. Since the must-carry rules prevent the cable operator editorial discretion over their own networks, the Supreme Court held in Turner I that the rules interfered with the First Amendment rights of cable operators.

But the Communications Act Allows Internet Filtering

Internet regulation advocates faced huge problem, though. Unlike other expansions of FCC authority into media, Congress was not silent about regulation of the Internet. Congress announced a policy in the 1996 update to the Communications Act that Internet access providers should remain “unfettered by State and Federal regulation.”

Regulation advocates dislike Section 230 because of its deregulatory message and because it expressly allows Internet access providers to filter the Internet.

Professor Yochai Benkler, in agreement with Lawrence Lessig, noted that Section 230 gives Internet access providers editorial discretion. Benkler warned that because of 230, “ISPs…will interject themselves between producers and users of information.” Further, these “intermediaries will be reintroduced not because of any necessity created by the technology, or because the medium requires a clearly defined editor. Intermediaries will be reintroduced solely to acquire their utility as censors of morally unpalatable materials.”  

Professor Jack Balkin noted likewise that “…§ 230(c)(2) immunizes [ISPs] when they censor the speech of others, which may actually encourage business models that limit media access in some circumstances.” 

Even the FCC acknowledges the consumer need for curated services and says in the Open Internet Order that Title II providers can offer “a service limited to offering ‘family friendly’ materials to end users who desire only such content.”

While that concession represents a half-hearted effort to bring the Order within compliance of Section 230, it simply exposes the FCC to court scrutiny. Allowing “family friendly” offers but not other curated offers is content-based distinction. Under Supreme Court RAV v. City of St. Paul, “[c]ontent-based regulations are presumptively invalid.”  Further, the Supreme Court said in US v. Playboy, content-based burdens must satisfy the same scrutiny as content-based bans on content. 

Circuit Split over the First Amendment Rights of Common Carriers

Hopefully the content-based nature of the Title II regulations are reason enough for the Supreme Court to take up an appeal. Another reason is that there is now a circuit split regarding the extent of First Amendment protections for common carriers.

The DC Circuit said that the FCC can prohibit content blocking because ISPs have been labeled common carriers.

In contrast, other courts have held that common carriers are permitted to block content on common carrier lines. In Information Providers Coalition v. FCC, the 9th Circuit held that common carriers “are private companies, not state actors…and accordingly are not obliged to continue…services of particular subscribers.” As such, regulated common carriers are “free under the Constitution to terminate service” to providers of offensive content. The Court relied on its decision a few years earlier in  Carlin Communications v. Mountain States Telephone and Telegraph Company that when a common carrier phone company is connecting thousands of subscribers simultaneously to the same content, the “phone company resembles less a common carrier than it does a small radio station” with First Amendment rights to block content. 

Similarly, the 4th Circuit in Chesapeake & Potomac Telephone Co. v. US held that common carrier phone companies are First Amendment speakers when they bundle and distribute TV programming, and that a law preventing such distribution “impairs the telephone companies’ ability to engage in a form of protected speech .” 

The full DC Circuit will be deciding whether to take up the Title II challenges. If the judges decline review, the Supreme Court would be the final opportunity for a rehearing. If appeal is granted, the First Amendment could play a major role. The Court will be faced with a choice: Should the Internet remain “unfettered” from federal regulation as Congress intended? Or is the FCC permitted to perpetuate itself by bringing legacy media regulations to the online world?

The success of the Internet and the modern digital economy was due to its open, generative nature, driven by the ethos of “permissionless innovation.” A “light-touch” policy regime helped make this possible. Of particular legal importance was the immunization of online intermediaries from punishing forms of liability associated with the actions of third parties.

As “software eats the world” and the digital revolution extends its reach to the physical world, policymakers should extend similar legal protections to other “generative” tools and platforms, such as robotics, 3D printing, and virtual reality.

In other words, we need a Section 230 for the “maker” movement.

The Internet’s Most Important Law

Today’s vibrant Internet ecosystem likely would not exist without “Section 230” (47 U.S.C. § 230) of the Telecommunications Act of 1996. That law, which recently celebrated its 20^th anniversary, immunized online intermediaries from onerous civil liability for the content and communications that travelled over their electronic networks.

The immunities granted by Section 230 let online speech and commerce flow freely, without the constant threat of legal action or onerous liability looming overhead for digital platforms. Without the law, many of today’s most popular online sites and services might have been hit with huge lawsuits for the content and commerce that some didn’t approve of on their platforms. It is unlikely that as many of them would have survived if not for Section 230’s protections.

For example, sites such as eBay, Facebook, Wikipedia, Angie’s List, Yelp, and YouTube all depend on Section 230 immunities to shield them from potentially punishing liability for the content that average Americans post to those sites. But Section 230 protects countless small sites and services just as much as those larger platforms and it has been an extraordinary boon to online commerce and speech.

Extending Immunities to Other General-Purpose Technologies: 3 Models

To foster generativity and permissionless innovation for the next wave of tech entrepreneurs, it may be necessary to immunize some intermediaries (i.e., platform providers or device manufacturers) from punishing forms of liability, or at least to limit liability in some fashion to avoid the chilling effect that excessive litigation can have on life-enriching innovation. Specifically, they should be immunized from liability associated with the ways third-parties use their platforms or devices to speak, experiment, or innovate.

“The past ten years have been about discovering new ways to create, invent, and work together on the Web,” noted Chris Anderson in his book Makers: The New Industrial Revolution. “The next ten years will be about applying those lessons to the real world.” But that can only happen if we get public policy right.

Thus, the creators of newer general-purpose technologies may need to receive certain limited immunizations from liability for the ways third-parties use their devices. If troublemakers use general-purpose technologies to do harm—i.e., cybersecurity violations, privacy invasions, copyright infringement, etc.—it is almost always more sensible to hold those problematic users directly accountable for their actions.

The other approach—holding those intermediaries accountable for the actions of third parties—will discourage innovators from creating vibrant, open platforms and devices that could facilitate new types of speech and commerce. Therefore, an embrace of permissionless innovation requires a rejection of such middleman deputization schemes.

There are three different existing immunity models we might consider applying to emerging general-purpose technologies.

Model #1: Section 230 & online services

The first model, of course, is Section 230 itself.  Section 230 stipulated that it is the policy of the United States “to promote the continued development of the Internet and other interactive computer services and other interactive media,” and “to preserve the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.” To accomplish that, the law made it clear that, “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”

Since implementation of Section 230 two decades ago, courts have generally read this immunity fairly broadly, so much so that some critics have argued that 230’s scope has been enlarged well beyond congressional intent. Even if that is true, I believe that has been a net positive (excuse the pun) and that it is not only wise to preserve that sweeping immunity but extend it to other technologies and sectors.

Model #2: Firearm manufacturing

Another immunization model can be found in the Protection of Lawful Commerce in Arms Act of 2005 (Pub. L. No. 109-92, 119 Stat. 2095). Although “lawsuits alleging negligent distribution plagued the firearm industry until 2005,” the Protection of Lawful Commerce in Arms Act “effectively ended the ‘gun tort’ era,” notes Peter Jensen-Haxel. The law did so by granting gun manufacturers immunities for such legal actions. (It would seem that, by extension, those who use 3D printers to create firearms will also be immunized from civil actions.)

Importantly, unlike Section 230, which provided broad immunity by default to all online platforms, the Protection of Lawful Commerce in Arms Act applied to manufactures/sellers that fit into the certain qualifications (i.e., they get immunity if they comply with certain licensing rules, record keeping requirements, etc.). This tension between broad versus targeted immunity will become the subject of debate for emerging general-purpose technologies as scholars and policymakers contemplate optimal default liability rules.

Model #3: Vaccines

A final legal immunization model comes, ironically, from the world of medical immunizations. As part of the National Childhood Vaccine Injury Act of 1986 (42 U.S.C. §§ 300aa-1 to 300aa-34), Congress created The National Vaccine Injury Compensation Program, “after lawsuits against vaccine companies and health care providers threatened to cause vaccine shortages and reduce U.S. vaccination rates, which could have caused a resurgence of vaccine preventable diseases.”

As described by the U.S. Department of Health and Human Services, the program, “is a no-fault alternative to the traditional legal system for resolving vaccine injury petitions.” Thus, those suffering injuries from vaccines are able to seek compensation from this program instead of having to sue vaccine companies.

As Avery Johnson of the Wall Street Journal noted in 2009 article about the program, “A spate of lawsuits against vaccine makers in the 1970s and 1980s had caused dozens of companies to get out of the low-profit business, creating a public-health scare. The strategy worked and the public health implications have been sizable. Vaccines have driven huge reductions — and in the case of smallpox, for instance, complete eradications — of major childhood diseases.”

This model is obviously very different than Section 230 and the Protection of Lawful Commerce in Arms Act in that it includes a government-created compensation fund provided as an alternative to civil lawsuit remedies. In all likelihood, such a compensation fund would not be necessary for new general-purpose “maker” technologies or sectors.

Nonetheless, this model could, perhaps, have some relevance for certain narrow classes of those technologies. For example, 3D-printed medical devices might be one area where it would make sense to exempt from liability the creators of 3D printers and the platforms over which 3D printer blueprints are distributed. But if there is significant resulting harm from some of those devices or plans, it remains unclear how compensation would work and who would be picking up the tab for it. The National Vaccine Injury Compensation Program offers one potential answer, although it may not be wise to craft such a consumer-funded or taxpayer-supported program for other reasons. Even if creating a government-run compensation fund was eventually seen as a good idea, we cannot determine how big the fund should be until some actual harms occur.

Three Sectors to Cover

Next, we should consider which sectors or technologies should be eligible for such immunities.

I wish it was possible to craft some sort of “General-Purpose Technology Immunization Act” that would shield such platforms and technologies from onerous liability associated with third-party uses. Realistically, however, it is not likely such a broad-based regime could achieve political traction. There would just be too many opposing forces. Moreover, there may be some unique distinctions between technologies and sectors which necessitate specialized legal regimes.

In any event, I believe a good case can be made for adopting some sort of legal immunity regime for three specific technologies: Robotics, 3D printing, and immersive technology (i.e., virtual reality and augmented reality).

Robotics

Ryan Calo, professor of law at the University of Washington School of Law, has done important work on the law of robotics and he has suggested that such legal immunities may need to be extended to this field. In his 2011 Maryland Law Review article on “Open Robotics,” Calo made his case as follows:

To preempt a clampdown on robot functionality, Congress should consider immunizing manufacturers of open robotic platforms from lawsuits for the repercussions of leaving robots open.  Specifically, consumers and other injured parties should not be able to sue roboticists, much less recover damages, where the injury resulted from one of the following: (1) the use to which the consumer decided to put the robot, no matter how tame or mundane; (2) the nonproprietary software the consumer decided to run on the robot; or (3) the consumer’s decision to alter the robot physically by adding or changing hardware. This immunity would include lawful and unlawful uses of the robot. (p. 134) . . . The immunity I propose is selective: Manufacturers of open robots would not escape liability altogether. For instance, if the consumer runs the manufacturer’s software and the hardware remains unmodified, or if it can be shown that the damage at issue was caused entirely by negligent platform design, then recovery should be possible. The immunity I propose only applies in those instances where it is clear that the robot was under the control of the consumer, a third party software, or otherwise the result of end-user modification. Because this issue will not always be easy to prove, we should expect litigation at the margins. I am thus arguing for a compromise position: A presumption against suit unless the plaintiff can show the problem was clearly related to the platform’s design. (p. 136)

I find this entirely convincing and I also believe Calo is wise to begin with robotics as the first target for such legal immunization because such technologies are already being widely manufactured and deployed today.

These liability questions are already being widely debated, for example, in the field of autonomous systems and driverless cars in particular. I’d like to believe that the common law would sort out these things fairly quickly and that an efficient liability regime would emerge from autonomous technologies in short order.

Alas, because America lacks a “loser pays” rule, a perverse incentive exists for overly-zealous trial lawyers to file an avalanche of lawsuits at the first sign of any problem. This could significantly hamper the development of autonomous technologies, which have the potential to immediately decrease the staggering death toll associated with human error behind the wheel. Therefore, it may be necessary for Congress to craft some sort of limited immunity regime for autonomous technology makers to ensure that the development of these potential life-saving technologies is not discouraged by the looming threat of perpetual litigation.

3D Printing

3D printing would be my second choice for a general-purpose technology that should be covered by some sort of intermediary immunity model.

In a forthcoming law review article for the Minnesota Journal of Law, Science & Technology, Adam Marcus and I argue that “the manufacturers of 3D printing devices and the website operators hosting blueprints for 3D-printed objects may need to be protected from liability to avoid chilling innovation. In this sense, a ‘Section 230 for 3D printing’ might be needed.”

We discuss three specific ways that 3D printers could be used by third-parties in such a way that existing laws or regulations are implicated and someone might seek to bring action against the manufacturers of 3D printers or 3D printing marketplaces, like Shapeways or Thingiverse. These cases involve things like 3D-printed prosthetics, which could raise policy concerns at the Food and Drug Administration, and 3D-printed toys or sculptures, which could present intellectual property issues.

But perhaps the most interesting case study for liability purposes will be 3D-printed firearms, which are already raising a great deal of controversy. Marcus and I argue, once again, that “the proper focus of regulation should remain on the user and uses of firearms, regardless of how they are manufactured.” And because, as already noted, the Protection of Lawful Commerce in Arms Act immunizes gun manufacturers from legal liability for third-party actions, it would seem logical that the law’s protections would extend to 3D-printed firearms. Moreover, Section 230 itself (and perhaps also the First Amendment) might also apply to 3D printing design schematics that appear on various websites or 3D printing marketplaces.

Generally speaking, Marcus and I argue, “imposing liability on third parties—sites hosting schematics, search engines, and manufacturers of devices—seems neither workable nor wise. There exists a broad spectrum of general-purpose technologies that can be used to facilitate criminal activity,” we note, such as cars, computers, or paper printers. But we don’t blame those intermediaries when those technologies are used by third parties in criminal acts. The same principle should apply to 3D printers.

Things get more complicated when intellectual property issues are brought into the debate. In an important 2014 article, “Patents, Meet Napster: 3D Printing and the Digitization of Things,” Deven R. Desai and Gerard N. Magliocca sketched out the potential case for some sort of limited immunity as it pertains to patent infringement and 3D printing. “An obstacle to the growth of 3D printing that Congress should consider addressing is that individuals who engage in that activity are strictly liable if they infringe a patent,” they note, but they continue on to add that:

Exempting personal 3D printing from patent infringement without undermining other aspects of the regulatory scheme will not be easy. It would not be a good idea for Congress to create a fair use exception for all patents or make infringement an intentional tort, as those changes would sweep too far. Targeting 3D printing itself is a possibility, but in that case the legislation would have to distinguish between personal and commercial activity, as there is no rationale for saying that all 3D printing leading to patent infringement, including what Fortune 500 firms do, should be permitted. Drawing that kind of line with a substantive legal standard, though, will generate years of litigation and may not effectively separate the good from the bad. One alternative, should Congress opt to give personal 3D printing some immunity, would be to set a relatively high minimum amount-in-controversy for federal jurisdiction over any [patent] infringement claims involving this technology. (p. 1717)

Getting this balance right will be tricky, yet essential. “Patent law and industries that rely on patents will have to adapt to this new environment or face potential obsolescence,” Desai and Magliocca correctly conclude.

Immersive Technology

A final sector we might eventually want to apply some sort of intermediary immunity model to is immersive technology. “Immersive technology” refers to services that currently utilize wearable devices (such as a head-mounted display or headset) to let users explore virtual worlds, virtual objects, or hologram-like projections. Immersive technology can be separated into two different, but related groups: virtual reality (VR) and augmented reality (AR).

These technologies are still in the cradle, but many companies are already developing VR and AR technologies for both entertainment and professional uses. As they gain more widespread usage, immersive technologies could raise some policy issues, including concerns about privacy, intellectual property (ex: who owns certain “experiences”), and potentially even worries about distraction and addiction.

It would not be surprising, therefore, if some critics begin advocating greater regulation of, or liability for, VR and AR intermediaries. If that happens, policymakers will need to consider immunizing them from the threat of lawsuits or else innovation will die in these sectors.

Conclusion

Following the general logic of permissionless innovation, and understanding the importance of keeping intermediaries free of punishing liability for what others might do with their general-purpose technologies and platforms, the proper focus of regulation should remain on the user and uses of those technologies.

Accordingly, policymakers should craft a “Section 230 for the maker movement” by adopting legal protections for robotics, 3D printing, and immersive technology. At the same time, we should seek out better solutions—legal and otherwise—to the old problems that might persist or new ones that might come about due to the use of these new devices and platforms. But we should not let hypothetical worst-case scenarios and concerns about future technologies lead us down a path where intermediaries are “deputized” or hit with punishing liability for downstream actions by third parties.

Note#1 : This is a preliminary sketch of a law review article I would eventually like to write entitled, “A Section 230 for the “Makers” Movement: Extending Section 230 Immunities to Robotics, 3D Printing & Virtual Reality.” Toward that end, I welcome suggestions for (a) which general-purpose technologies deserve some sort of immunization, and also (b) what other legal immunity regimes exist that we could learn from. Please forward any ideas you might have along to me.

Note #2: My thanks to Adam Marcus and Christopher Koopman for their helpful suggestions on this essay.

The FCC’s Open Internet Order is long and complex and the challenge to it is likewise difficult to untangle. The agency regularly engages in ad hoc rulemaking that results, per Judge Posner, in “unprincipled compromises of Rube Goldberg complexity among contending interest groups viewed merely as clamoring suppliants who have somehow to be conciliated.” The Open Internet Order is no exception and therefore faces several legal vulnerabilities.

In my view, the soft underbelly of the Order is the agency’s position that ISPs are not First Amendment speakers. While courts are generally very deferential to agencies, they are not deferential on constitutional questions. Further, the court panel (two Democrat appointees, one Republican appointee), unfortunately, was not in the carriers’ favor. The major carriers, however, have focused their arguments on whether the agency should receive deference in classifying Internet access as a telecommunications service.

That said, it’s possible the major carriers could get at least a partial win with their arguments. That likelihood is increased because Alamo Broadband and Dan Berninger raised the First Amendment problems with the Order. Given the strength of the First Amendment arguments, the Court might shy away from reaching the issue of whether ISPs are speakers. Below, some thoughts on the moments during oral arguments that surprised me and what went according to predictions.

The Unexpected

A receptive ear in Judge Williams re: the First Amendment arguments. (Good for: ISPs) The First Amendment arguments went better than I’d expected. Alamo and Berninger’s counsel, Brett Shumate, argued the First Amendment issues well and had good responses for skeptical questions. Shumate found a receptive ear in Judge Williams, who seemed to understand the serious First Amendment risks posed by the Order. Williams repeatedly brought up the fact that MetroPCS a few years ago tried to curate the Internet and provide its customers free YouTube, only to face resistance from the FCC and net neutrality activists.

The other two judges were more skeptical but Shumate corrected some misconceptions. The biggest substantive objection from Srinivasan, who sounded the most skeptical of the First Amendment arguments, was that if the Court reaches the First Amendment issues, it has determined that the FCC has reasonably classified Internet access as a common carrier service. He suggested that this means the First Amendment issues mostly disappear. No, Shumate explained. Congress and the FCC can call services whatever they want. They could declare Google Search or Twitter feeds a common carrier service tomorrow and that would have zero effect on whether filtering by Google and Twitter is protected by the First Amendment. Tatel asked whether Section 230’s liability protections suggest ISPs are common carriers and Shumate corrected that misconception, a subject I have written on before.

A major FCC concession that ISPs have to option to change their offerings and escape common carrier regulation. (Good for: ISPs) Title II advocates are spinning the terse First Amendment exchanges as a victory. I’m not convinced. The reason the arguments didn’t generate more heat was because the FCC lawyer made a huge concession at the outset: ISPs that choose to filter the Internet are not covered by the Open Internet Order.

FCC lawyer: “If [ISPs] want to curate the Internet…that would drop them out of the definition of Broadband Internet Access Service.” Judge Williams: “They have that option under the Order?” FCC lawyer: “Absolutely, your Honor. …If they filter the Internet and don’t provide access to all or substantially all endpoints, then they drop out of the definition of [BIAS] and the rules don’t apply to them.”

This admission seriously undermines the purposes of the Order. The FCC is stating outright that ISPs have the option to filter and to avoid the rules. That seems to mean that Comcast’s Stream Internet protocol television service, where it is curating streaming TV programs, is not covered by the rules. If Facebook’s Free Basics or a similar service launched in the US giving free, limited access to the Web, that is not covered by the Order. Finally, this means that the many broadband packages that offer family-friendly filtering are outside of the FCC’s rules. It’s not clear how much remains to be regulated since all ISPs reserve the right to filter content and each filters at least some content.

Judge Tatel directing most questioning. (Good for: wash) Many view Judge Tatel as the “swing vote” but I was surprised at the relative quiet from Williams and Srinivasan. Tatel was the most inquisitive, by my listening. He was much more skeptical of some of the FCC’s arguments regarding interconnection than I expected but also more skeptical of the First Amendment arguments than I expected.

Little discussion of Chevron Step 0. (Good for: FCC) Many on the free-market side wanted to make this case about Chevron Step 0 and the notion that Title II is too economically and socially significant to warrant deference. Unfortunately, at oral argument there was very little discussion of Chevron Step 0.

The Expected

Focus on agency discretion. (Good for: FCC) The judges generally seem to see this as a straightforward Chevron case and the questions focused on Chevron Step 1, whether there is ambiguity in the statute about “offering telecommunications” for the FCC to interpret. As expected, the FCC did fairly well in their arguments because these technical issues are very hard to untangle.

On Chevron Step 2, whether the reinterpretation of “telecommunications service” to include Internet access was reasonable, the US Telecom attorney was strong. He leaned heavily on the fact that in Section 230, which amends the Communications Act, Congress announces a national policy that the Internet and specifically Internet access services, should remain “unfettered by Federal regulation.” That would seem to preclude the FCC from using, at the very least, its most powerful regulatory weapon–common carriage–against Internet access providers. Even if “telecommunications service” is ambiguous, he stated, it was unreasonable to include Internet access in that definition.

Focus on whether mobile broadband can be properly classified under Title II. (Benefit: ISPs) As many commentators have noted, the idea that the traditional phone network and the mobile broadband network can be classified as the same interconnected network is far-fetched. Each judge seemed very skeptical of the FCC’s argument and Tatel suggested there was a lack of adequate notice.

Srinivasan pointed out that striking down the wireless rules and maintaining the wireline rules would mean that using the same tablet in different areas of your house would lead to different regulatory treatment, depending on whether you’re on the cellular broadband network or Wifi. Title II supporters think this is pretty clever gotcha but communications law already abounds with seemingly absurd FCC- and court-created legal distinctions. (The FCC invents its own absurd distinction and offers vastly different regulatory treatment for DNS operated by an ISP v. DNS operated by literally anyone else.)

Conclusion

Predictions about major regulatory cases are notoriously difficult. I’ve read (and made) enough predictions about big court cases to know that prognosticators almost always get it wrong. If that’s the case, at least consider one thought-provoking outcome: the rules are largely struck down because the FCC provided inadequate notice on most of the major issues of classification.

If the rules, in contrast, were sustained under Chevron and judged to have had adequate notice, the Court would likely need to confront the First Amendment issues. I don’t think Tatel and Srinivasan, especially, want to rule on these hard constitutional questions. The judges must know the Supreme Court has, as Prof. Susan Crawford says, an “absolutist approach” to the First Amendment that protects speakers of all kinds. Sustaining the rules means the FCC risks a loss on First Amendment grounds on appeal that would nearly eliminate the ability of the FCC to regulate the Internet. For that reason, and because of the notice problems, the Court may strike down the rules on notice and comment grounds, thereby preserving the ability of the FCC to take a fourth bite at the apple.

As FCC Commissioner Jessica Rosenworcel said of the Internet, “It is our printing press.” Unfortunately, for First Amendment purposes, regulators and courts treat our modern printing presses — electronic media — very differently from the traditional ones. Therefore, there is persistent political and activist pressure on regulators to rule that Internet intermediaries — like social networks and search engines — are not engaging in constitutionally-protected speech.

Most controversial is the idea that, as content creators and curators, Internet service providers are speakers with First Amendment rights. The FCC’s 2015 Open Internet Order designates ISPs as common carriers and generally prohibits ISPs from blocking Internet content. The agency asserts outright that ISPs “are not speakers.” These Title II rules may be struck down on procedural grounds, but the First Amendment issues pose a significant threat to the new rules.

ISPs are Speakers Courts and Congress, as explained below, have long recognized that ISPs possess editorial discretion. Extensive ISP filtering was much more common in the 1990s but still exists today. Take JNet and DNet. These ISPs block large portions of Internet content that may violate religious principles. They also block neutral services like gaming and video if the subscriber wishes. JNet offers several services, including DSL Internet access, and markets itself to religious Jews. It is server-based (not client-based) and offers several types of filters, including application-based blocking, blacklists, and whitelists. Similarly, DNet, targeted mostly to Christian families in the Carolinas, offers DSL and wireless server-based filtering of content like pornography and erotic material. A strict no-blocking rule on the “last mile” access connection, which most net neutrality proponents want enforced, would prohibit these types of services.

The extensive filtering these ISPs do is not as common as it once was, but they illustrate the reality that ISPs do have editorial discretion and they do exercise it. For that reason, scholars have pointed out for nearly a decade that any meaningful no-blocking rule compels speech from ISPs and other Internet platforms. ISPs would be forced to transmit content that they object to.

Therefore, the Title II rules, especially a no-blocking rule, may not survive a First Amendment challenge. Title II proponents frequently assert that ISPs are “dumb pipes” and must be prohibited from exercising any editorial control over what is transmitted from consumers. Net neutrality activists around the globe, for instance, want to prohibit Internet.org, a free app jointly produced by Facebook and wireless carriers, because it “blocks” access to content that is is not selected to be included within the package.

To avoid scrutiny from a court, the FCC will need to show that ISPs resemble common carriers like telephone companies and FedEx that — though transmitting speech — don’t have editorial discretion and have essentially no First Amendment rights. However, if ISPs instead resemble electronic media like cable TV companies or search engines that exercise editorial control over transmitted content, the Title II rules represent compelled speech and will receive significant court scrutiny.

Internet Platforms and Curation A court holding that ISPs are speakers for First Amendment purposes is important because Internet-based, curated distribution is replacing traditional ways consumers accessed news and media — bookstores, newsstands, broadcast radio, cable TV. The nature of “publication” and “speakers” has changed rapidly in a short time because information is much more accessible in the Internet age. Clay Shirky notes that the traditional formula of “Filter, then publish,” has been replaced with “Publish, then filter.”

While the nature of media changed, many still want to regulate the intermediaries. Google’s algorithm is a common target for regulation. Politico recently published a piece calling for search engine regulation because of Google’s ostensible ability to sway close elections through opaque algorithm tweaks. Many online companies and media companies likewise want to regulate the order in which Search results appear, an effort gaining traction in Europe. Further, some academics and activists would like to extend neutrality rules to Twitter’s and Facebook’s curation of user streams.

Fortunately, those efforts would likely fail in the US because Internet intermediaries receive constitutional protection. As Eugene Volokh persuasively argues, “search engines are speakers” and regulations affecting Google’s algorithms must withstand First Amendment scrutiny. Law professor Jonathan Zittrain advises regulatory caution and notes that “content curators…have a First Amendment right to present their content as they see fit.” ISPs likewise have an existing right (seldom exercised) to curate and filter content. As net neutrality supporter Harold Feld says, if the FCC doesn’t classify ISPs as common carriers, “nothing requires your ISP to deliver [Internet content]. If Comcast decides I am evil and blocks [my website], they can do it.”

Courts call this ability to filter, anachronistically, “editorial discretion.” In this new media world of content abundance, “curation” better represents what defines protected speech because most of the actual messages transmitted originate from third parties. Certainly, print organizations have substantial editorial control over what is published, but radio and TV is less controlled and newer media platforms display a wide range of editorial controls. Many of the most important modern speakers in this “publish, then filter” environment are curators. New media, like aggregators, cable companies, search engines, and ISPs, often use an intentional, semi-automated, iterative process to decide what content to omit and what to transmit.

Constitutional protection of curators means regulators likely cannot force a Christian cable operator to carry Cinemax. Likewise, Apple can continue to block apps containing Confederate flags or nudity in the App Store. Even though ISP filtering of content may be a blunter tool, like a WISP operated by Christians or Jews that blocks websites for its religious users, or like Internet.org that transmits only select Internet content, the First Amendment protects that ability to tailor content.

The Dumb Pipes Myth Readers of law review articles from net neutrality advocates and of the Open Internet Order are left with the false impression that ISPs are passive transmitters — dumb pipes — that never block content. Title II supporters have to maintain this facade because if they suggest that ISPs exercise editorial control, the no-blocking rules trigger First Amendment scrutiny. Fortunately for First Amendment purists, there is sufficient evidence of ISPs filtering content to raise serious questions about the constitutionality of the Open Internet Order. Internet service providers can and do engage in filtering, and some curate content in ways that are much more intentional than First Amendment-protected speech activity like cable TV distribution and Google search results.

There are, as mentioned, small ISPs like JNet and Clean Internet that provide Internet access marketed to religious users. In addition to its wireline service, DNet is also a wireless ISP and advertises that they filter pornography and other content. Sprint, one of the Big Four national wireless carriers, last year offered social media plans and parental controls, including whitelists and blacklists, through its Virgin Mobile subsidiary.

I’ve posed this question to Title II advocates several times — Aren’t religious ISPs engaged in First Amendment-protected speech? To date, none have denied it and seem content to pretend server-based filtering by wired and wireless access providers doesn’t exist. Susan Crawford, for instance, wrote a law review piece about ISPs’ purported lack of First Amendment protection. She noted, correctly, that if ISPs were engaged in editorial decisions about filtering content, that would pose a problem for the FCC regulations. However, she abruptly discontinued a Twitter back-and-forth with me when I pointed out she omitted instances of religious ISPs exercising the editorial discretion she fears.

It’s not just smaller ISPs that filter. AT&T, for instance, like nearly every major ISP and Web company, reserves the right in its acceptable use policy to pull down content “that is determined by AT&T to be obscene, indecent, hateful, malicious, racist, defamatory, fraudulent, libelous, treasonous, excessively violent or promoting the use of violence or otherwise harmful to others.” This is not idle language. Groups like the Anti-Defamation League, for instance, knowing that they have enforceable acceptable use policies actively lobby and persuade ISPs and Web companies to remove content from anti-Semites and groups like the KKK.

Title II proponents like to point out that the large ISPs engage in relatively little of the curation and filtering that I’ve described. Therefore, they reason, ISPs are common carriers. I’m not persuaded. That distinction appears immaterial considering the FCC made no such distinction in its Open Internet Order. Further, large ISP reluctance to offer, say, family-friendly Internet packages is entirely predictable considering the FCC has for a decade chilled that exercise of free speech through ham-fisted attempts at net neutrality enforcement, merger conditions, and punitive fines.

Congress Intended to Encourage ISPs to Filter Content with Section 230 The FCC faces another obstacle to its conclusory determination that ISPs are not speakers. The 2015 Open Internet Order largely adopts net neutrality proponents’ First Amendment arguments and concludes that ISPs “serve as mere conduits for the messages of others, not as agents exercising editorial discretion.” However, Section 230 of the Communications Decency Act has a Good Samaritan provision that suggests ISPs are speakers, a view that several courts have endorsed.

For better or for worse, Section 230 makes ISPs and other Internet platforms that primarily rely on third-party content “super speakers.” Internet platforms can exercise editorial control and get all the benefits of being a speaker, like First Amendment protection, yet are immunized from many of the burdens, like liability for distributing online defamation and libel.

Why did Congress take this dramatic step in the mid-1990s? Quite simply, the drafters wanted to encourage ISPs to continue to block offensive content online. At that time, ISPs and bulletin board operators like Prodigy marketed themselves as family friendly and (inconsistently) blocked content that Prodigy administrators judged to be in bad taste. Because of this editorial discretion, in 1995 Prodigy faced costly liability in a defamation suit for defamatory statements one of its users posted, in a case called Stratton Oakmont v. Prodigy. Because of Stratton Oakmont and a few similar cases, Internet intermediaries faced two undesirable options to avoid liability for transmitted content:

1. become conduits and exercise no editorial control — thereby leaving even offensive content online; or
2. constantly police Internet content and take down all questionable material.

Congress disliked both options and quickly responded with Section 230 protections in 1996 to protect Internet-based distributors from becoming mere conduits. The statute protects “interactive computer services,” which includes, “specifically a service or system that provides access to the Internet….” Congress, therefore, preserved ISPs’ editorial role in cleaning up the Internet.

Several court cases recognize that ISPs and other Internet platforms exercise editorial discretion and are not mere conduits. As the district court said in the 1998 case Blumenthal v. Drudge, 230’s protections serve “as an incentive to Internet service providers to self-police the Internet for obscenity and other offensive material.” Similarly, the 9th Circuit Court of Appeals noted in the 2008 Roommate.com case that through 230, “Congress sought to [allow ISPs] to perform some editing on user-generated content….” A 10th Circuit decision states that “Congress clearly enacted § 230 to forbid the imposition of publisher liability on a service provider for the exercise of its editorial and self-regulatory functions.” Finally, as the Fourth Circuit said in Zeran v. America Online, Section 230 “forbids the imposition of publisher liability on a service provider for the exercise of its editorial and self-regulatory functions.”

At the very minimum, we have several courts saying that ISPs exercise editorial discretion — directly contradicting the assertions by Title II proponents and the FCC. Further, not only can ISPs filter content, it is a practice that Congress wished to encourage. The Title II rules chill those editorial functions, and that is a problem for the FCC.

Finally, while not a Section 230 case, a majority of the Supreme Court has tacitly endorsed Congress’ view that ISPs are speakers that can and should serve as curators of online content. In Ashcroft v. ACLU, the Court cited ISP filtering favorably as an alternative to unconstitutional provisions of the Child Online Protection Act. Taking these cases together, it’s unlikely a court will sustain the net neutrality advocate view that ISPs are not speakers and cannot engage in filtering.

The 2015 Open Internet Order and its Content-Neutrality Problem Section 230 poses an additional problem for the FCC. Most courts construe Section 230(c)(2) broadly in terms of what services are protected and the kinds of liability providers are immunized from. A broad reading may protect ISPs from FCC regulations that restrain ISP filtering abilities.

The FCC’s hastily-written Title II order seems to recognize this but reveals some internal tensions in an effort to allow some Section 230-type ISP filtering. For example, while the Order says ISPs are not First Amendment speakers, the Order appears to accept the premise that Section 230 restrains agency action. Specifically, the Order likely allows ISPs to block offensive content and offer family-friendly packages because the FCC permits, in paragraph 220, ISPs to block “traffic that is unwanted by end users.” The FCC cites to the portions of the 2010 Open Internet Order that allowed Internet access packages that block, specifically, pornographic content. In those referenced portions of the 2010 Order, the FCC cites Section 230 and expressly states that the agency will not impose liability for good-faith actions by ISPs to restrict harassing and offensive content.

This exception to the no-blocking rule, if it is indeed an exception, puts the FCC in a bind when defending its no-blocking rule against First Amendment challenges. As an initial matter, this exception that allows ISPs to actively block the content described in Section 230 suggests FCC acknowledgement that ISPs are exercising editorial control and engaged in protected speech.

Further, if the FCC interprets this “unwanted traffic” exception narrowly, the agency would allow ISPs to block only lewd, harassing, and violent material (that is, the material specifically mentioned in Section 230), but would penalize ISPs for blocking, say, political, religious, and entertainment content. Such a distinction means that the Title II rules are not content-neutral regulations and therefore likely to be struck down on First Amendment grounds. In the words of the Supreme Court, courts will “apply the most exacting scrutiny to regulations that…impose differential burdens upon speech because of its content.” If, on the other hand, the FCC interprets “unwanted traffic” broadly in a content-neutral manner, the rules allow widespread filtering of Internet content, undermine the FCC’s assertion that ISPs are not speakers, and eviscerate the entire purpose of the Title II rules.

Before concluding, one aside is probably necessary: This position — that ISPs have a constitutional right to filter Internet packages for consumers — is often misrepresented and maligned by net neutrality advocates. To be clear, that ISPs or other Internet platforms have a First Amendment right to select which online content to transmit does not mean anticompetitive blocking or throttling of, say, Netflix is permissible. That is illegal. It likewise does not mean ISPs can promise subscribers access to Internet content and subsequently block that content. That is also illegal. This is a narrower claim: The First Amendment protects ISPs’ prerogative to transparently offer kid-friendly and other filtered Internet packages like Internet.org.

The future of the media is Internet-based and content is increasingly curated. It’s important that regulators don’t deprive ISPs of their First Amendment rights because ISPs represent the canary in the coalmine. First Amendment protection has generally been a one-way ratchet that’s expanded free speech protections for several decades and has been used successfully to fight regulations that affect speech. A setback for the First Amendment would encourage media access activists to seek more regulation for new media. The “gatekeeper” theory used to justify regulation of ISPs is just the warmed-over “scarcity” rationale for the Fairness Doctrine and other restraints on speech. The calls to regulate Twitter, Facebook, and Google algorithms and Internet television will grow even louder if courts accept the FCC’s First Amendment analysis in the Open Internet Order. Fortunately, the Open Internet Order suffers from several deficiencies and there is a good chance courts will again remind regulators to keep their hands off the Internet.

Article was originally posted on Plain Text on September 3.

It’s been over five years since Congress passed major legislation addressing copyright protection, but this hasn’t stopped copyright owners from achieving real progress in securing their expressive works. In cooperation with private-sector stakeholders, rights holders have made several deals aimed at combating copyright infringement and channeling consumer demand for original content toward legitimate outlets. These voluntary agreements will be the subject of a hearing this afternoon (9/18) before the House Judiciary Committee’s Subcommittee on Courts, Intellectual Property and the Internet. This panel marks the latest in a series of hearings the committee launched earlier this year to review the Copyright Act, much of which dates back to 1976 or earlier.

Copyright consensus may sound like an oxymoron, especially in the wake of last year’s bruising legislative battle over SOPA and PIPA. But in reality, there’s no shortage of common ground when it comes to copyright protection. Despite all the controversy that surrounds the issue, copyright isn’t so much a “conflict of visions”, to borrow from Thomas Sowell, but a conflict of tactics, as I argued earlier this year on Cato Unbound.

Indeed, with some notable exceptions, most scholars, business leaders, and policymakers accept that government has a legitimate and important role in securing to inventors and creators the fruits of their labors“. Unsurprisingly, the devil is in the details, where genuinely tough questions arise regarding the government’s proper role in policing the Internet for copyright violations. Should the law hold online intermediaries accountable for their users’ infringing acts? What remedies should the law afford rights holders whose works are unlawfully distributed all over the Internet, often by profit-generating foreign actors?

Although Congress has struggled mightily with these questions, business leaders from a variety of sectors have worked together to devise several approaches to the problem of copyright infringement that go above and beyond the Copyright Act. Perhaps most notably, in February 2013, a coalition of five major ISPs and several trade associations representing filmmakers and artists announced the launch of the Copyright Alert System (“CAS”). Administered by the Center for Copyright Information, the CAS aims to educate users about copyright law—and deter them from violating it—by delivering Copyright Alert notices to ISP subscribers found to be sharing infringing files on peer-to-peer networks.

It’s too early to render a verdict on the CAS’s effectiveness, but as data accumulates in coming months and years, researchers will surely examine how the system has impacted user behavior. Similar approaches to infringement by ISP subscribers have been tried in other countries such as France—albeit on a mandatory, not voluntary, basis—and several studies have found that these so-called “graduated response” systems have indeed reduced infringement in nations where they’ve been implemented. But other studies have reached the opposite conclusion, so more research is needed in this area. Whether or not CAS succeeds, however, experimentation involving novel approaches to copyright protection is crucial for the future of creative expression, as is experimentation among business models to monetize content.

Speaking of voluntary approaches to copyright protection, Google last week unveiled a report describing its anti-piracy efforts. As Google’s Fred von Lohmann explained:

[W]e are releasing a report, “How Google Fights Piracy,” bringing together in one place an overview of the programs, policies, and technologies we have put in place to combat piracy online.

The report discusses how Google penalizes websites that receive a high percentage of DMCA takedown notices in Google’s search results, hopefully thereby directing users toward legitimate sources of content. It explains the “Content ID” system pioneered by YouTube, which enables rights holders to identify potentially infringing videos posted to the site, and gives copyright owners the choice to monetize such videos in lieu of removing them altogether. And the report points out that in 2012, Google voluntarily disabled ad service to 46,000 websites dedicated to infringement. Check out the full report for much more information on these efforts and many others that Google has taken to better secure copyrights—and for another perspective, check out a MPAA-commissioned study released today critiquing the role that Internet search engines play in helping users find infringing websites.

After today’s hearing, I’ll have more thoughts on the state of voluntary cooperation to protect copyrights, and on the debate about whether file lockers and search engines ought to do more to combat infringement.

Andrew Orlowski of The Register (U.K.) recently posted a very interesting essay making the case for treating online copyright and privacy as essentially the same problem in need of the same solution: increased property rights. In his essay (“‘Don’t break the internet’: How an idiot’s slogan stole your privacy“), he argues that, “The absence of permissions on our personal data and the absence of permissions on digital copyright objects are two sides of the same coin. Economically and legally they’re an absence of property rights – and an insistence on preserving the internet as a childlike, utopian world, where nobody owns anything, or ever turns a request down. But as we’ve seen, you can build things like libraries with permissions too – and create new markets.” He argues that “no matter what law you pass, it won’t work unless there’s ownership attached to data, and you, as the individual, are the ultimate owner. From the basis of ownership, we can then agree what kind of rights are associated with the data – eg, the right to exclude people from it, the right to sell it or exchange it – and then build a permission-based world on top of that.”

And so, he concludes, we should set aside concerns about Internet regulation and information control and get down to the business of engineering solutions that would help us property-tize both intangible creations and intangible facts about ourselves to better shield our intellectual creations and our privacy in the information age. He builds on the thoughts of Mark Bide, a tech consultant:

For Bide, privacy and content markets are just a technical challenges that need to be addressed intelligently.”You can take two views,” he told me. “One is that every piece of information flowing around a network is a good thing, and we should know everything about everybody, and have no constraints on access to it all.” People who believe this, he added, tend to be inflexible – there is no half-way house. “The alternative view is that we can take the technology to make privacy and intellectual property work on the network. The function of copyright is to allow creators and people who invest in creation to define how it can be used. That’s the purpose of it. “So which way do we want to do it?” he asks. “Do we want to throw up our hands and do nothing? The workings of a civilised society need both privacy and creator’s rights.”  But this a new way of thinking about things: it will be met with cognitive dissonance. Copyright activists who fight property rights on the internet and have never seen a copyright law they like, generally do like their privacy. They want to preserve it, and will support laws that do. But to succeed, they’ll need to argue for stronger property rights. They have yet to realise that their opponents in the copyright wars have been arguing for those too, for years. Both sides of the copyright “fight” actually need the same thing. This is odd, I said to Bide. How can he account for this irony? “Ah,” says Bide. “Privacy and copyright are two things nobody cares about unless it’s their own privacy, and their own copyright.”

These are important insights that get at a fundamental truth that all too many people ignore today: At root, most information control efforts are related and solutions for one problem can often be used to address others. But there’s another insight that Orlowski ignores: Whether we are discussing copyright, privacy, online speech and child safety, or cybersecurity, all these efforts to control the free flow of digitized bits over decentralized global networks will be increasingly complex, costly, and riddled with myriad unintended consequences. Importantly, that is true whether you seek to control information flows through top-down administrative regulation or by assigning and enforcing property rights in intellectual creations or private information.

Let me elaborate a bit (and I apologize for the rambling mess of rant that follows).

Parallels in Debates over Copyright & Privacy Protection

In several essays here over the past few years I have attempted to draw parallels between the battles over protecting digital copyright and online privacy, as well as battle over online safety/speech and cybersecurity. Here are a few of those essays in case you’re interested in seeing the evolution of my thinking about this:

• When It Comes to Information Control, Everybody Has a Pet Issue & Everyone Will Be Disappointed
• And so the IP & Porn Wars Give Way to the Privacy & Cybersecurity Wars
• SOPA & Selective Memory about a Technologically Incompetent Congress
• Privacy as an Information Control Regime: The Challenges Ahead
• Isn’t “Do Not Track” Just a “Broadcast Flag” Mandate for Privacy?

In those essays I have argued that a combination of selective morality and wishful thinking are at work in the information policy world these days. In essence, people hate Internet regulation… until they love it! Here’s how I summarized that fact during the debate over SOPA:

… conservatives rush out and breathlessly denounce each and every effort to impose Net neutrality regulation because of the danger of empowering an already over-zealous bunch of bumbling bureaucrats at the FCC. (And I agree with them.) Yet, with their next breath many conservatives praise SOPA even though it also empowers government to muck with the inner workings of the Internet. Some of those conservatives are also turning a blind eye to the growing appetite of the defense/security community to meddle with the Net’s architecture in the name of avoiding any number of non-catastrophes. Meanwhile, the liberals decry SOPA and want it stopped at all costs. There’s never been a copyright protection measure they liked, of course, but each time one pops up we hear them claim that our analog era Congress is not well-positioned to be designing industrial policy schemes for the Internet. (And I generally agree with them.) But most liberals do a complete 180 whenever online privacy or Net neutrality regulations are the subject of congressional inquiry. Suddenly, the cyber-oafs in Congress are considered veritable technocratic philosopher kings who we should trust to guard our cyber-freedoms to lead us to the digital promised land.

Again, it’s both selective morality and wishful thinking. It’s selective morality in that some folks think certain values are sacrosanct and deserving of a “by-any-means-necessary” enforcement attitude, yet they are often just as likely to denounce similar information control efforts when it comes to issues or values they don’t give a damn about.  And it is wishful thinking in that you can’t run around insisting that “information wants to be free” in some contexts but then express outrage when something that you want to bottle up turns out to “just want to be free” as well!

But the important takeaway here is that, consistent with what Orlowski argues, I believe that online copyright and privacy are essentially the same problem: It’s an information control problem.

Potential Costs of Control

Once you start thinking about Internet policy debates as a single issue — namely, information control — you can begin to investigate the potential costs of control in a somewhat more objective fashion. Of course, challenging issues remain:

1. Which method of control should we choose? On one hand, there are many varieties of administrative regulation, technical infrastructure controls, and device mandates. On the other hand, there are property rights and liability / tort schemes. And there are many hybrid enforcement models, such as increasingly popular “co-regulation” models, government standard-setting, and “nudging” of system defaults. Each method will entail different costs and trade-offs.
2. What metric(s) should we use when attempting to determine whether the benefits of control exceed the costs? Ask any advocate of information control about whether the costs might exceed the benefits of regulation for their pet issue and they will typically suggest that either (a) there are no costs or that (b) the benefits dwarf any costs that may exist. But all too often the benefits they identify are extremely subjective and amorphous in character (“privacy,” “safety,” and “security” are hard to quantify, after all) while the costs are very real and increasingly substantial.

In my view, these practical questions are increasingly the most interesting issues to explore in the field of cyberlaw and digital economics. We can debate the normative or ethical considerations until we’re all blue in the face and ready to rip each other’s heads off, but I am less and less interested in such squabbles. Instead, I keep coming back to the question of how we’ll go about controlling info flows and how much effort and resources it makes sense to expend in pursuit of each of the values identified above. Some of the specific considerations I find myself asking in every paper I write these days include:

(A) Will the proposed form of information control tie us up in the courts forever, lead to increasingly onerous and unworkable liability norms, and end up yielding outrageous litigation costs?

(B) Will the proposed form of information control require a significant increase in regulatory bureaucracy? How many levels of government will need to be involved in the proposed enforcement scheme? How many new offices and officials will need to be empowered in the hope of achieving some measure of control?

(C) What are the alternatives to the proposed form of information control? Are there less costly or less restrictive means of addressing the concern in question? For example, education and empowerment effort are often an effective way to address many online safety and digital privacy concerns. Can we use those methods in conjunction with social norms, public pressure, self-regulation, informal contracting, and other methods to address these and other concerns?

For me, the costs associated with the A & B are increasing so rapidly that I almost always default to C as the better approach. Importantly, although A & B will be less onerous or costly when the solution is of the increased property-ization variety than of the administrative regulation variety, that does not mean property rights-based solutions for information are costless. Indeed, I increasingly find myself concluding that C solutions are more cost-effective even compared to increased property rights.

Practical Advice Once You Accept the Increasing Costs & Complications of Control

At this point, readers may be thinking: “Wait a minute, this dude is just some kooky libertarian who doesn’t want any form of information control, so he’s just trying to rationalize anarchy here.” No, I’m not. I certainly favor less control across the board than most people, but I also understand that there are times, at the margin, when some forms of “control” are necessary. But my views on the wisdom of control are heavily influenced by the costs of control. The costs of control — broadly defined — are a key factor in every cost-benefit analysis I do related to the wisdom of Net regulation and information control methods — even when one of those methods is increased “property-ization.” And because I have come to believe that those costs are going up and that most information control efforts will not work well in practice, I have boiled down my advice on this front to two simple principles:

1. Choose your info control battles wisely. Figure out where the most serious harms or threats lie and then target the info control solution accordingly and forget about the rest. For example, in child safety debates, that would mean going after child porn rings but leaving run-of-the-mill adult porn alone entirely. In copyright, it would mean nailing the largest commercial mass piracy sites but accepting a certain amount of casual sharing. In the field of personal info, it means singling out health and financial information and data for special protections and likely giving up on most other forms of info control. And so on. In essence, these are where the greatest potential harms lie that most people would consider intolerable. As you move further away from such issues, the case for control becomes harder and harder and the costs will almost certainly exceed the benefits.
2. Have a good backup plan in mind when those info control plans fail anyway. That backup plan should generally be based on education, empowerment, coping strategies, and resiliency. Again, these are the “C” solutions mentioned above. [I developed this model more robustly in the second half of this recent paper.] This approach won’t be perfect but it will likely be what you’ll end up relying on anyway, so you better start thinking about plowing more resources into this alternative approach even while you’re trying to devise info control mechanisms.

Let me just say a brief word to my market-oriented friends who are dismayed by my inclusion of property rights in the mix of “information control” efforts. I’m a big believer in the importance of property rights in many contexts, but context does matter. More specifically, physicality matters. It is easy to create property rights in tangible goods and almost always right to do so. Property rights in intangible ideas and creations raise special issues, however. Because ideas are non-rivalrous and have public good qualities, it makes property-ization more complicated and less effective. Property rights in facts can also come into conflict with other values and more well-established rights, especially freedom of speech and expression.

On the privacy front, Eugene Volokh made this point in his famous 2000 law review article, “Freedom of Speech, Information Privacy, and the Troubling Implications of a Right to Stop People from Speaking About You,” when he noted that, “The difficulty[with] the right to information privacy — the right to control other people’s communication of personally identifiable information about you — is a right to have the government stop people from speaking about you. And the First Amendment (which is already our basic code of “fair information practices”) generally bars the government from “control[ling the communication] of information,” either by direct regulation or through the authorization of private lawsuits.” That doesn’t mean free speech values should always trump privacy values, but denying this tension is just plain silly. If you want to propertytize all personal information, then you better be prepared to explain how that plays out in practice. How far are you prepared to go to ban the dissemination of facts? Would you place prior restraint on the press to accomplish it? Would you ban a historian from writing a biographies that reveal intimate facts about the subject? Would you shut down all the online sites and services that rely on a certain amount of personal information to fuel their free offerings?

Likewise, copyright law was far more effective in the analog age when we were still pressing music on vinyl and plastic. As soon as digitization become widespread, it was pretty much game over for traditional copyright law and now we are off and running with all sorts of convoluted and increasingly costly regulatory regimes. It’s not that I don’t want these some of these schemes to work — I’ve been a long-time copyright defender — but, again, the practicality of control simply must be considered here. I am not will to “pay any price, bear any burden” in defense of protecting intellectual property rights even as I remain outraged by the staggering amount of free-riding at work every single second of the day on the Internet. So, adopting the framework I outlined about, we might try targeted solutions to go after the biggest of those freeloaders — commercial mass piracy hubs — but we should generally avoid the sort of ham-handed technical control methods we saw in SOPA and other fights, like the broadcast flag battle among others. But, generally speaking, property rights just aren’t going to work as well in this space going forward. I’ve come to believe that the best hope lies in massive consolidation of content and conduit. In other words, pipe and device owners need to buy out all the content-creating industries and just embed a small fee in their monthly services to cross-subsidize content. This is essentially a private collective licensing solution and it is not unprecedented. Nor is it perfect. It will be very leaky. Plenty of piracy will still take place. But it will probably offer creators a better chance of finding a sustainable revenue stream than the current system does. The old copyright system that served them and us so well is dying and they had better start thinking of alternatives like this. Of course, antitrust law may never allow it, so I could be wasting my breath here. (Just look at all the grief that antitrust officials both here and abroad are giving Apple and eBook sellers for working together even though that it probably the best scheme devised in recent memory to sustain publishing in an age of mass piracy. Policymakers should be encouraging more of that sort of thing, not punishing it.)

An Uncertain Future

So, to wrap up… I can imagine a future in which both heavy-handed, top-down info control efforts and property / liability solutions are failing almost universally because of the ubiquitous, instantaneous, quicksilver-like flow of information across decentralized digital networks. Some utopians will argue that such a world will be better in every way than the one we live in today. I do not share such hyper-optimism. While I believe that, on balance, the free flow if information generally benefits society, I also understand how it creates enormous angst and intractable challenges for many. It’s a world in which copyright is a hollow shell of its former self that offers creators very little protection for their expressive works. And it’s a world in which personal privacy is harder to safeguard with each passing day because no matter how hard we try to property-tize facts about ourselves, that enforcement model simply breaks down at some point or becomes socially and economically intolerable. As with copyright, efforts to property-tize personal information will lose the battle against data sharing. As computer scientist Ben Adida argued in his essay, “(Your) Information Wants to be Free,” “unfortunately, information replication doesn’t discriminate: your personal data, credit cards and medical problems alike, also want to be free. Keeping it secret is really, really hard.”

Indeed, and it is growing harder by the day. Contrary to what Orlowski suggests, therefore, this isn’t a simple engineering problem. I wish it were as easy as he suggests to build “permissions-based markets” because they could have real benefits for individuals and society. But it is most certainly not that simple. It is far more costly and complicated than ever to devise workable information control schemes on one hand and “permissions-based” property rights schemes on the other. In some cases, I might still be willing to try the latter, but unlike Orlowski, I just don’t place much faith in the success of the endeavor.

In the ongoing debate over SOPA, PIPA, and rogue websites legislation, most commentators have focused on what Congress should and shouldn’t do to combat these sites. Less attention, however, has been paid to the underlying assumption that these rogue websites represent a public policy problem. While no one has defended websites that defraud consumers by deceptively selling them fake pharmaceuticals and other counterfeit goods, many consumers who frequent “rogue websites” do so for the express purpose of downloading copyright infringing content.

As Julian Sanchez explains over on Cato-at-Liberty, how the latter category of rogue websites (including The Pirate Bay and, until last week, MegaUpload) affects the U.S. economy and social welfare is hotly contested in the economic literature:

[I]t’s become an indisputable premise in Washington that there’s an enormous piracy problem, that it’s having a devastating impact on U.S. content industries, and that some kind of aggressive new legislation is needed tout suite to stanch the bleeding. Despite the fact that the [GAO] recently concluded that it is “difficult, if not impossible, to quantify the net effect of counterfeiting and piracy on the economy as a whole,” our legislative class has somehow determined that . . . this is an urgent priority. Obviously, there’s quite a lot of copyrighted material circulating on the Internet without authorization, and other things equal, one would like to see less of it. But does the best available evidence show that this is inflicting such catastrophic economic harm—that it is depressing so much output, and destroying so many jobs—that Congress has no option but to Do Something immediately? Bearing the GAO’s warning in mind, the data we do have doesn’t remotely seem to justify the DEFCON One rhetoric that now appears to be obligatory on the Hill. The International Intellectual Property Alliance . . . actually paints a picture of industries that, far from being “killed” by piracy, are already weathering a harsh economic climate better than most, and have far outperformed the overall U.S. economy through the current recession.

Julian makes several great points, and his essay is well worth reading in its entirety.

Nevertheless, in my view, rogue websites dedicated to the infringement of U.S. copyrights pose a public policy problem that merits not only serious congressional attention, but also prompt (albeit prudent) legislative action. While I’m relieved that the flawed SOPA and PIPA bills seem unlikely to pass in their current forms, I also think it would be unwise for Congress to dither on rogue sites legislation for years in search of “credible data” about how such sites impact our economy.

Why am I urging policymakers act without “all the facts?” Two reasons. First, I’m quite skeptical that we’ll obtain anything resembling dispositive data on the question of how rogue websites impact consumer welfare in the foreseeable future. Countless academics have spent years seeking to understand how often consumers download content on rogue websites, how frequently consumers substitute unlawful content for the lawful kind, and the extent to which copyright infringement indirectly benefits creators by inducing greater overall content consumption. Yet reliable data on these topics remains the stuff of dreams.

Second, the ease with which U.S. consumers can and do access near-perfect infringing copies of movies, songs, television shows, and video games gives rise to a reasonable presumption that we’d probably be better off if Congress were to throw up at least some carefully-constructed roadblocks to obstruct rogue sites. That’s because if such roadblocks are erected, the consumers most likely to shift from unlawful to lawful consumption of content are also the same consumers who are most likely to benefit social welfare (and the U.S. economy) if they pay more for the content they value and enjoy.

Imagine two hypothetical “pirates” (or users who frequently infringe on copyrighted works, if you prefer less loaded terminology). Pirate #1 is a broke college sophomore with a subsidized ultra-fast broadband connection and eons of spare time on his hands. While this pirate lacks the disposable income to pay for content at virtually any price, he’s perfectly willing to spend hours on end sitting hunched over a laptop in his dorm room scouring various Web forums for links to his favorite TV shows and movies, most of which are available unlawfully on cyberlockers, Bittorrent, Usenet, etc.

Pirate #2 is a 30-something, tech-savvy mid-career IT professional with plenty of disposable income. Even though he owns a Blu-ray player and could afford to buy or rent several discs per month, he instead opts to download Blu-ray image files on his 50Mbps Verizon FiOS connection and watches them on a laptop hooked up to his high-def television. Using his Mastercard, he spends $10 a month to subscribe to a popular Seychelles-based content search website that enables him to find picture-perfect movie rips in seconds. Although he has the means to pay for content, he sees no reason to bother with physical discs, DRM, and platform restrictions given that pirated content is so much cheaper, and virtually as accessible. While he ultimately purchases some of the content he acquires unlawfully, attending the occasional live concert and theatrical performance, he only does so occasionally. He has few moral qualms about his behavior; with millions of other consumers paying for the content he enjoys, what difference can one more legitimate purchaser make? (Julian correctly observes that some individuals who “sample” music through illicit outlets ultimately spend more money on artists because they’re more likely to attend live performances. However, given the growing prevalence of free and lawful sources of music “samples,” and considering that piracy’s effects on creators of other types of content (e.g. movies, TV shows, video games) is far less ambiguous, the “file sharing actually benefits artists” hypothesis is hardly persuasive).

Turning back to the issue of rogue sites legislation, a law that serves only to make it impossible for Pirate #1 (and the millions in America like him) to access infringing content won’t do anybody much good. Content creators won’t get paid more, as Pirate #1 has no money, while the aggregate utility society derives from artists’ expressive works will decrease. Instead of enjoying movies and music acquired unlawfully, Pirate #1 will simply find another, presumably inferior, way to spend his free time. It’s a no-win situation.

But a law that makes it impossible for Pirate #2 (and the millions like him) to access infringing content would almost certainly benefit content creators — and society at large. No longer able to download movies, TV shows, and video games illegally, Pirate #2 might consume less overall content, but he’ll also pay for a lot more lawfully-acquired content. He’ll spend less of his disposable income on goods and services other than content, meaning some legitimate businesses will experience a decline in revenue. But since Pirate #2’s overall spending habits will more closely match his true consumption preferences, society’s aggregate resources will likely end up being allocated more efficiently than before .

The virtue of a “follow the money” approach to rogue websites is that it’s likely to curb piracy by users like Pirate #2, who are already willing and able to pay for legitimate content. Users who have a credit card and use it to pay for infringing content — or for services that facilitate access to infringing content — presumably have at least some disposable income to spend on expressive works. While rogue websites legislation is likely to leave many, if not most, websites that facilitate piracy unaffected, disabling U.S. payment services from doing business with a handful of especially popular offshore piracy sites will frustrate users. Many of these users will simply seek out alternatives, but some users will give up and “go legitimate.” By driving piracy further underground, such a law might cause users like Pirate #1 to spend more of their relatively worthless time seeking out infringing content. But this is the Internet we’re talking about; the determined user will find what he seeks, no matter the roadblocks lawmakers throw up.

Whether a targeted law aimed at combating offshore rogue sites’ revenue sources would, on net, measurably benefit the U.S. economy is far from certain. But even a law that has greater-than-even odds of improving aggregate social welfare by the equivalent of a few hundred million dollars amounts to a step in the right direction. In a world of uncertainty, we all make decisions with harshly limited knowledge every day. All else equal, making highly-informed decisions is vastly superior to educated guesses, but educated guesses are often the best feasible option.

In an ideal world, of course, Congress would be focused on far more crucial legislative priorities than combating rogue websites, such as solving the entitlement mess, fixing America’s overly litigious legal system, reining in the ever-growing regulatory state, and even reforming the Copyright Act to reduce the insanely long term of copyright protection. But given that both the House and Senate Judiciary Committees, which handle copyright legislation, seem more focused on undermining our liberty and prosperity than on enhancing it — from data retention to employment verification mandates to the PATRIOT Act renewal — passing a consensus rogue websites bill may be the best of all feasible outcomes this session of Congress.

If lawmakers act swiftly but carefully — holding a handful of additional hearings, focusing on crafting legislation that Silicon Valley can tolerate (if not embrace), and emphasizing a transparent process — there may still be hope for prudent rogue websites legislation this session. And that could be a good thing.

Over at TIME.com, I write that while Congress mulls an Internet blacklist in SOPA, there are efforts underway to reengineer parts of the Net to make communications more decentralized and censorship-proof. These include distributed and decentralized DNS systems, currencies, and social networks, as well as attempts to circumvent ISPs using mesh networking.

It’s not a certainty that these projects will all succeed. Most probably won’t. Yet these far-out efforts serve as proof-of-concept for a censorship-resistant Internet. Just as between Napster and BitTorrent there was Gnutella and Freenet, it will take time for these concepts to mature. What is certain is the trend. The more governments squeeze the Internet in an attempt to control information, the more it will turn to sand around their fingers.

Read the whole thing here.

The Stop Online Piracy Act (SOPA), a controversial bill before the House of Representatives aimed at combating “rogue websites,” isn’t just about criminal, foreign-based sites that break U.S. intellectual property laws with impunity. Few dispute that these criminal websites that profit from large-scale counterfeiting and copyright infringement are a public policy problem. SOPA’s provisions, however, extend beyond these criminal sites, and would potentially subject otherwise law-abiding Internet intermediaries to serious legal risks.

Before moving forward with rogue websites legislation, it’s crucial that lawmakers take a deep breath and appreciate the challenges at stake in legislating online intermediary liability, lest we endanger the Nozickian “utopia of utopias” that is today’s Internet. The unintended consequences of overbroad, carelessly drafted legislation in this space could be severe, particularly given the Internet’s incredible importance to the global economy, as my colleagues have explained on these pages (1, 2, 3, 4, 5, 6)

To understand why SOPA could be a game-changer for online service providers, it’s important to understand the simmering disagreement surrounding the Digital Millennium Copyright Act (DMCA) of 1998, which grants certain online service providers a safe harbor from liability for their users’ copyright infringing actions. In exchange for these protections, service providers must comply with the DMCA’s notice-and-takedown system, adopt a policy to terminate users who repeatedly infringe, and meet several other conditions. Service providers are only eligible for this safe harbor if they act to expeditiously remove infringing materials upon learning of them. Also ineligible for the safe harbor are online service providers who turn a blind eye to “red flags” of obvious infringement.

The DMCA does not, however, require providers to monitor their platforms for infringing content or design their services to facilitate monitoring. Courts have held that a DMCA-compliant service provider does not lose its safe harbor protection if it fails to act upon generalized knowledge that its service is used for many infringing activities, in addition to lawful ones, so long as the service provider does not induce or encourage users’ infringing activities.

Defenders of the DMCA safe harbor argue that it’s helped enable America’s Internet-based economy to flourish, allowing an array of web businesses built around lawful user-generated content — including YouTube, Facebook, and Twitter — to thrive without fear of copyright liability or burdensome monitoring mandates.

Conversely, some commentators, including UCLA’s Doug Lichtman, argue that the DMCA inefficiently tips the scales in favor of service providers, to the detriment of content creators — and, ultimately, consumer welfare. Pointing to a series of court rulings interpreting the safe harbor’s provisions, critics argue that the DMCA gives online intermediaries little incentive to do anything beyond the bare minimum to stop copyright infringement. Critics further allege that the safe harbor has been construed so broadly that it shields service providers that are deliberately indifferent to their users’ infringing activities, however rampant they may be.

What does SOPA have to do with all of this? Buried in the bill’s 78 pages are several provisions that run a very real risk of effectively sidestepping many of the protections conferred on online service providers by the DMCA safe harbor.

Section 102

Section 102 of SOPA empowers the Attorney General to seek a court order against an allegedly infringing foreign website. Such a court order would, if granted, effectively deny the site access to payment processors, ad networks, and even parts of the domain name system. Under § 102, a foreign, U.S.-directed website is deemed a “foreign infringing site” if:

[T]he owner or operator of such Internet site is committing or facilitating the commission of criminal violations [involving illegal copyright infringement, counterfeiting, or theft of trade secrets] and the Internet site would . . . [therefore] be subject to seizure in the United States . . . if such site were a domestic Internet site.

The part about websites “subject to seizure in the United States” refers to 18 U.S.C. § 2323, which states among other things that “[p]roperty subject to forfeiture” includes:

Any property used, or intended to be used, in any manner or part to commit or facilitate the commission of [criminal copyright or trademark infringement].

This definition of a “foreign infringing site” is enormously troubling. Note the absence of any requirement of actual or constructive knowledge on the part of the site operator, let alone criminal intent. Under § 102, a foreign website built around user-generated content may be deemed an “infringing site” simply because its server has facilitated the criminally infringing acts of a single user — even if the site operator neither induced nor knew of the user’s unlawful activities. While an innocent foreign site operator might eventually be able to persuade a court to vacate an order deeming it a “foreign infringing site,” SOPA imposes an astonishingly low burden on the Attorney General of showing that a site is a “foreign infringing site.” If the bill is enacted as is, foreign websites that contain any user-generated content had better watch out.

SOPA proponents defend § 102 by pointing out that its definition of infringing sites comes straight out of the 2008 PRO-IP Act, which established the aforementioned civil forfeiture provision in 18 U.S.C. § 2323. But this statute’s constitutionality is currently being challenged in federal court by a team of attorneys that includes Stanford law professor and copyright guru Mark Lemley. The law’s breadth raises serious First Amendment concerns since it permits ex parte seizures of entire outlets of speech (e.g., websites) simply because the outlet has been used in some unlawful manner. SOPA may be based on existing law, but why should Congress extend this overbroad provision of the PRO-IP Act to encompass an even broader range of websites? If anything, lawmakers should revisit PRO-IP and narrow its applicability to sites intentionally operated for the purpose of committing or facilitating criminal infringement. Via Techdirt, even Floyd Abrams, a constitutional scholar who represents content companies that strongly back SOPA, conceded in a recent letter to Congress that unanswered questions remain regarding the constitutionality of 18 U.S.C. § 2323.

Section 103

The next section of SOPA, Section 103, isn’t any better. This section provides for private rights holders to seek court orders against U.S.-directed websites — including domestic sites — to deny them access to U.S. payment processors and ad networks. Section 103 deems a website “dedicated to theft of U.S. property” if any of the following conditions are met:

1. [The site] is primarily designed or operated for the purpose of, has only limited purpose or use other than, or is 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 [copyright infringement, circumvention of copyright protection systems, or trademark infringement]; or
2. [The site operator] is taking, or has taken, deliberate actions to avoid confirming a high probability of the use of the . . . site to carry out acts that constitute [copyright infringement or the circumvention of copyright protection systems]; or
3. [The site operator] operates the . . . site with the object of promoting, or has promoted, its use to carry out acts that constitute [copyright infringement or the circumvention of copyright protection systems], as shown by clear expression or other affirmative steps taken to foster infringement.

The first prong of this definition encompasses any website that “has only limited purpose or use other than . . . engag[ing] in, enabl[ing], or facilitat[ing]” copyright infringement, circumvention of copyright protection systems, or trademark infringement. This language comes from 17 U.S.C. § 1201, also known as the DMCA anti-circumvention provisions. Just how “limited” of non-infringing uses must a site have to meet this definition? It’s hard to say. As Rob Pegoraro cheekily observed in a recent Roll Call op-ed, “‘[l]imited’ is one of those wonderfully elastic words — notice the ever-longer yet still ‘limited’ copyright terms granted to artists and creators?” This section of SOPA would be more clear if it relied on the “capable of substantial non-infringing uses” test originally articulated by the U.S. Supreme Court in its famous 1984 Betamax opinion, Sony Corp. v. Universal City Studios, Inc., which has since been interpreted by numerous federal courts in copyright infringement cases.

The second prong of the § 103 definition, which covers websites that take “deliberate actions to avoid confirming a [high probability of infringement],” is perhaps the most worrisome of the three prongs. This language appears to have been lifted directly from a 2011 U.S. Supreme Court decision,  Global-Tech Appliances, Inc. v. SEB S.A. In that case, a patent infringement lawsuit, the Court found the defendant liable for inducement on the grounds that it took willful steps to blind itself of the existence of the patent at suit. The Court held that “willful blindness” exists when (1) a defendant subjectively believes that there is a high probability that a fact exists; and (2) the defendant takes deliberate actions to avoid learning of that fact.

Note, however, that Section 103 omits the first prong of the Global Tech willful blindness test, the subjective belief element. This omission might simply be an oversight — or it could reveal the intent of the bill’s authors to cast aside the subjective knowledge standard (which currently applies to service providers in the context of knowledge for purposes of the DMCA) and replace it with an objective, “reasonable person” standard. If plaintiff bringing a SOPA action is only required to show that a website operator should have known of its users’ infringement from the perspective of a “reasonable” operator, and that the site’s operator acted in some manner that had the effect of contributing to its ignorance of infringing activities by users, a vast array of websites that currently enjoy the protections of the DMCA safe harbor may face significant new legal risks. After all, website operators make design decisions all the time that might foreseeably impact on their awareness (or lack thereof) of user’ potentially infringing activities. Who knows what sort of well-intentioned, albeit deliberate, decisions might amount to”avoiding confirming a high probability” of infringement?

As David Sohn of the Center for Democracy & Technology has pointed out, “[t]his seems like a backdoor way of imposing a monitoring obligation on any website that allows users to post content.”  Temple Law Professor David Post, writing at the Volokh Conspiracy, observed that the bill might make it a “violation of law to keep the prosecutors from ‘confirming’ that you’re violating the law — all the prosecutor has to show, to make you vanish from the Net, is that you’ve somehow tried to keep the prosecutor off of your website!”

Why SOPA Could Endanger the DMCA Safe Harbor

SOPA proponents have dismissed concerns that the bill would risk undermining the DMCA safe harbor. U.S. Register of Copyrights Maria Pallante, testifying in a House Judiciary Committee hearing on SOPA on November 16, told members of Congress that it was extremely unlikely that any actions brought under SOPA would impact websites otherwise shielded by the DMCA safe harbor. Techdirt reports that Viacom executive Stanley Pierre-Louis recently argued that SOPA would not “[expand] the scope of secondary liability claims and [diminish] DMCA protections,” noting that “[t]here is no rule that permits ‘willful blindness’ of obvious wrongdoing under U.S. law, and nothing in the DMCA or any other statute has been deemed to hold otherwise.”

Technically, Pallante and Pierre-Louis are correct; SOPA’s provision at 102(c)(2)(A)(iii) appears to leave existing doctrines of copyright liability vis-à-vis the DMCA safe harbor untouched.

In practice, however, SOPA has the potential to effectively usurp the DMCA safe harbor in important respects. If the bill is enacted, online service providers would face a new worst nightmare: being cut off from payment processors, ad networks, and possibly even Internet service providers. As Eric Goldman recently explained, if a “website goes offline because of cash flow problems caused by the cutoff attributable to a single UGC content item, all of the UGC on that website goes dark because of a single content item.”

To avoid such an outcome, website operators will likely do everything they can to avoid falling under SOPA’s definitions — even if that means going above and beyond the requirements of the DMCA safe harbor. While I’m all for websites voluntarily taking prudent and measured actions to combat unlawful user activities (e.g., YouTube’s Content ID system), there are good reasons to be very skeptical of any legislation that effectively imposes on site operators any duty or obligation to monitor, or facilitate the monitoring of, user activities.

Fair concerns have been raised by thoughtful commentators about the DMCA’s limitations and shortcomings. Those concerns deserve a serious examination in the halls of Congress, and perhaps may even merit some careful, targeted tweaks to the DMCA. But the extraordinary remedies provided contained in SOPA should be reserved for genuine rogue sites that willfully flout U.S. laws with impunity and are beyond the reach of U.S. law enforcement authorities. While there are U.S.-based websites out there that violate copyright and trademark laws, extraordinary remedies (such as “going after the money”) should not be the primary method of penalizing such sites. If a rights holder believes that a domestic website is infringing on its copyright or trademark, the proper means of obtaining recourse is to file a civil lawsuit and, when appropriate, seek injunctive relief. The U.S. Marshals Service is tasked with enforcing civil judgments and other court orders entered against domestic actors by federal courts, and parties may obtain writs of execution to order law enforcement intervention against American individuals or businesses that violate court orders.

We Have To Pass The Bill To Find Out What’s In It

Reasonable people read SOPA’s provisions in very different ways. For instance, Terry Hart, writing at Copyhype, has eloquently defended SOPA’s definitions, arguing that “[t]he actions that would subject a provider to SOPA’s provisions are the same ones that would subject it to a copyright infringement suit under existing law and are actions that would not be protected under DMCA safe harbors.” But while SOPA’s definitions are based largely on well-established, time-tested statutes and precedents, some of the language isn’t as clear-cut as it might seem at first glance, as I explain above.

As a result, it’s tough to predict how SOPA would actually impact online service providers. Federal judges vary widely in the methods they employ in attempting to interpret vague statutes. There is no such thing as stare decisis when it comes to statutory construction; some judges focus on the plain meaning of a statute’s language, while others pour through committee reports and hearing transcripts in hopes of divining the legislature’s true underlying intent.

With apologies to Nancy Pelosi, what this means is that we probably won’t know what’s in SOPA until it’s passed. Even then, only after years of costly litigation will the contours of the bill’s provisions likely begin to approach a state of clarity. Consider that the DMCA, now thirteen years old, continues to engender serious disagreement among federal courts to this day. (For instance, courts disagree on what it means for a service provider to take “volitional acts” that encourage users to engage in infringement.)

SOPA’s potential breadth is especially problematic given that its potential victims are small, entrepreneurial Internet start-ups that lack the resources to pay a team of lawyers to examine their operational decisions for potential SOPA violations. As leading high-tech venture capitalist Fred Wilson has argued, “venture capitalists will think more than twice about putting $3mm of early stage capital into startups if they know that the vast majority of the funds will go to pay lawyers to defend the companies instead of to hire engineers to create and build product.”

Lawmakers Should Tread Carefully

While combating rogue foreign websites that violate U.S. laws flagrantly and with impunity should be a priority for lawmakers, SOPA’s definitions and remedies are simply too broad and too vague in their current form. They would cast a cloud of legal uncertainty over America’s innovative, startup-driven Internet economy. It would be a grave mistake to grant such powerful new tools to Justice Department and rights holders and assume that federal trial judges will interpret SOPA’s provisions as narrowly as is necessary to ensure legitimate Internet companies do not suffer adverse effects.

The recent House Judiciary Committee hearing on SOPA made clear just how much work remains to be done to craft an effective but targeted approach to rogue sites. Serious questions remain unresolved — not only about SOPA’s impact of the DMCA safe harbor, but also about cybersecurity, due process and free speech. Additional hearings are needed to explore these important issues with Internet engineers, law professors, and venture capitalists. Marking up the legislation before the end of 2011 — as Chairman Lamar Smith desires, according to the National Journal — would be a serious mistake.

For more on SOPA and rogue websites legislation; see: 

• Ryan Radia, Preliminary Thoughts  on Stop Online Piracy Act (SOPA)
• Ryan Radia, Congress Should Amend SOPA to Address Cybersecurity, Due Process Concerns
• Ryan Radia, Congress takes another stab at combating Rogue Websites with the PROTECT IP Act
• Ryan Radia, Five Ways Congress Can Fix COICA Copyright Bill
• Joint Letter from Public Interest Groups to the House Judiciary Committee on the Stop Online Piracy Act
• Timothy B. Lee, ArsTechnica, The Stop Online Piracy Act: Big Content’s full-on assault against the Safe Harbor
• Letter from Sens. Ron Wyden, Rand Paul, Jerry Moran, and Maria Cantwell to Congressional Leadership objecting to the PROTECT IP Act
• U.S. Rep. Zoe Lofgren’s Opening Letter on SOPA
• Prof. Eric Goldman, Why I Oppose the Stop Online Piracy Act (SOPA)/E-PARASITES Act
• Prof. David Post, How About Occupy Hollywood?
• Fred Wilson, Protecting The Safe Harbors Of The DMCA And Protecting Jobs
• Terry Hart, SOPA: New remedies for existing liability
• Rob Pegoraro, Online Piracy Act is Copyright Overreach
• Center for Democracy & Technology, The Stop Online Piracy Act: Summary, Problems and Implications
• Sherwin Siy, Public Knowledge, House Version of Rogue Websites Bill Adds DMCA Bypass, Penalties for DNS Workarounds
• Techdirt’s SOPA coverage
• Parker Higgins, Electronic Frontier Foundation, What’s On the Blacklist? Three Sites That SOPA Could Put at Risk
• Corynne McSherry, Electronic Frontier Foundation, SOPA: Hollywood Finally Gets A Chance to Break the Internet
• Larry Downes, TechFreedom, SOPA: Hollywood’s latest effort to turn back time

This afternoon the Stop Online Piracy Act (H.R. 3261) was introduced by Rep. Lamar Smith of the House Judiciary Committee. This bill is a companion to the PROTECT IP Act and S.978, both of which were reported by the Senate Judiciary Committee in May.

There’s a lot some to like about the bill, but I’m uneasy about some quite a few of its provisions. While I’ll have plenty to say about this bill in the future, for now, here are a few preliminary thoughts:

• The bill’s definition of “foreign infringing sites” at p. 10 borrows heavily from 18 U.S.C. § 2323, covering any site that commits or facilitates the commission of criminal copyright infringement and would be subject to civil forfeiture if it were U.S.-based. Unfortunately, the outer bounds of 18 U.S.C. § 2323 are quite unclear. The statute, which was enacted only a few years ago, encompasses “any property used, or intended to be used, in any manner or part to commit or facilitate” criminal copyright infringement. While I’m all for shutting down websites operated by criminal enterprises, not all websites used to facilitate crimes are guilty of wrongdoing. Imagine a user commits criminal copyright infringement using a foreign video sharing site similar to YouTube, but the site is unaware of the infringement. Since the site is “facilitating” criminal copyright infringement, albeit unknowingly, is it subject to the Stop Online Piracy Act?
• Section 103 of the bill, which creates a DMCA-like notification/counter-notification regime, appears to lack any provision encouraging ad networks and payment processors to restore service to a site allegedly “dedicated to theft of U.S. property” upon receipt of a valid counter-notification and when no civil action has been brought. The DMCA contains a safe harbor protecting service providers who take reasonable steps to take down content from liability, but the safe harbor only applies if service providers promptly restore allegedly infringing content upon receipt of a counter notification and when the rights holder does not initiate a civil action. Why doesn’t H.R. 3261 include a similar provision?
• The bill’s private right of action closely resembles that found in the PROTECT IP Act. Affording rights holders a legal avenue to take action against rogue websites makes sense, but I’m uneasy about creating a private right of action that allows courts to issue such broad preliminary injunctions against allegedly infringing sites. I’m also concerned about the lack of a “loser pays” provision.
• Section 104 of the bill, which provides immunity for entities that take voluntary actions against infringing sites, now excludes from its safe harbor actions that are not “consistent with the entity’s terms of service or other contractual rights.” This is a welcome change and alleviates concerns I expressed about the PROTECT IP Act essentially rendering certain private contracts unenforceable.
• Section 201 of the bill makes certain public performances via electronic means a felony. The section contains a rule of construction at p. 60 that clarifies that intentional copying is not “willful” if it’s based on a good faith belief with a reasonable basis in law that the copying is lawful. Could this provision cause courts to revisit the willfulness standard discussed in United States v. Moran, in which a federal court found that a defendant charged with criminal copyright infringement was not guilty because he (incorrectly) thought his conduct was permitted by the Copyright act?

Over the weekend, Janet Morrissey of The New York Times posted an excellent article on the U.S. government’s continuing crackdown on Internet gambling. (“Poker Inc. to Uncle Sam: Shut Up and Deal“) Ironically, her article arrives on the same week during which PBS aired the terrific new Ken Burns and Lynn Novick documentary on the history of alcohol prohibition in the United States. It’s a highly-recommended look at the utter hypocrisy and futility of prohibiting a product that millions of people find enjoyable. If there’s a simple moral to the story of Prohibition, it’s that you can’t repress human nature–not for long, at least, and not without serious unintended consequences. Which is why Morrissey of the Times notes:

And so the poker world now finds itself in a situation many liken to Prohibition. America didn’t stop drinking when the government outlawed alcoholic beverages in 1919. And, in this Internet age, it won’t be easy to prevent people from gambling online, whatever the government says. “It’s a game of whack-a-mole,” says Behnam Dayanim, an expert on online gambling and a partner at the Axinn Veltrop & Harkrider law firm. “They’ve whacked three very large moles, but over time, more moles will pop up.”

Exactly right (except that it should be “whac” not “whack”! There’s no K in whac-a-mole.)  It reminds me of the paper that my blogging colleague Tom Bell penned back in 1999 for the Cato Institute with its perfect title: “Internet Gambling: Popular, Inexorable, and (Eventually) Legal.” As Tom noted back then:

Consumer demand and lost tax revenue will create enormous political pressure for legalization, which we should welcome if only for its beneficial policy impacts on network development and its consumer benefits. We should also welcome it for a more basic reason: as the Founders recognized, our rights to peaceably dispose of our property include the right to gamble, online or off.

Again, you can’t hold back human nature and the effort of millions to pursue happiness as they see fit. It was true of alcohol and it will be true of online gambling–eventually.

And although it represents the worst argument for legalization, Tom was right about the tax revenue benefits as a primary factor leading to legalization. As Morrissey notes in her Times piece:

Uncle Sam is leaving a lot of money on the table. Over 10 years, legal online gambling could generate $42 billion in tax revenue, according to the Congressional Committee on Taxation. An estimated 1.8 million Americans played online poker last year, and some make a living at it. Because of the legal issues in the United States, online card rooms typically base their computer servers elsewhere, in places like Costa Rica or, in the case of Full Tilt, in the Channel Islands.

It was the same story back during alcohol prohibition, of course. All the “money left on the table” was snatched up by foreign governments and organized crime, who were all too happy to satisfy the thirst Americans had. Some State governments have already realized this and are taking steps to partially legalize online gambling and get in on the action, as Morrissey reports:

Oddly enough, Internet gambling is already legal in the nation’s capital. Earlier this year, the District of Columbia became the first jurisdiction in the United States to legalize it. Officials there said they hoped the move would bring in $13 million to $14 million a year in tax revenue. But Washington may only be the start. Several bills now working their way through the House of Representatives would give online poker the run of the country.

Again, as Bell’s paper argued, it’s popular, inexorable, and it will eventually be fully legal. We just have to be patient while some lawmakers play through this latest silly experiment in legislating morality.