Stanford CIS

Remember That Time We Saved the Internet?

By Annemarie Bridy on

My Twitter feed tells me that today is the fifth anniversary of the day the Internet “went dark” in protest of the Stop Online Piracy Act (SOPA) and the Protect Intellectual Property Act (PIPA). For anyone who needs a reminder, SOPA and PIPA were pieces of copyright legislation touted by their proponents as necessary to prevent online piracy and to protect U.S. jobs in the film, television, and music industries. They required a range of online intermediaries—ad networks, search engines, ISPs, and payment processors—to prevent users from accessing “foreign infringing sites.” Their most controversial provisions required ISPs, through the (mal)functioning of the Domain Name System (DNS), to block access to “foreign infringing sites” that allegedly traffic illegally in copyrighted content.

Back in 2012, online opposition to the two bills mounted quickly as word spread that they required ISPs to blacklist websites. In an open letter to Congress, Google co-founder Sergey Brin and other prominent Internet entrepreneurs asserted that the legislation would give the U.S. government “power to censor the web using techniques similar to those used by China . . . and Iran.” Contributing to and marshaling web-roots resistance, the operators of Wikipedia made the unprecedented decision to shut down in protest for one day—January 18, 2012. In addition to Wikipedia, more than 100,000 Internet companies, including Google, Mozilla, Reddit, and I Can Has Cheezburger (of LOLcats fame), joined the one-day protest. Their forms of protest varied, but their message to their users and fans was unitary: “Petition your elected representatives to oppose these bills.” And petition their representatives people did—in droves. Google reported that 4.5 million people in one day signed its petition opposing SOPA and PIPA. Even the White House came out against them.

My Twitter feed also tells me, or at least strongly implies, that I should celebrate this day as a victory for the Open Internet. I would find that exhortation more compelling if I hadn’t spent the last five years documenting the piecemeal implementation of SOPA and PIPA’s site-blocking and site-starving provisions through privately negotiated “best practices” in anti-piracy and anti-counterfeiting. Many of the intermediaries who would have been compelled to act by SOPA have since voluntarily agreed to do so under pressure from members of Congress and the Executive Branch. Under a series of "voluntary agreements," which the EFF calls “shadow regulation,” operators of major online advertising and payment networks have agreed to choke the supply of revenue to alleged pirate site operators in response to notices from copyright and trademark holders, without even the minimal judicial process that SOPA and PIPA would have required. If you’re interested in reading more about these agreements, I’ve written about them at length here and here.

Just last year, two registry operators for new generic Top Level Domains (gTLDs) in the DNS agreed to a domain-blocking protocol that designates the Motion Picture Association of America (MPAA) as a “trusted notifier” for purposes of identifying “pirate sites.” Under the agreement with the MPAA, participating registry operators can cancel or suspend entire second level domains in response to right holder complaints, without first undertaking anything that one would recognize as a fair alternative dispute resolution process. The MPAA responds to concerns about lack of fair process in this program by assuring those who inquire that the program is targeting only “the worst of the worst.” I believe they’re sincere when they say this. But their “trust us” logic flouts the most basic principles of fairness in a system of justice. Fairness isn’t about identifying complainants we think we can trust and then more or less automatically giving them the remedies they want against the parties they identify as bad guys. Fairness is about testing complaints in a neutral forum with a qualified adjudicator, clear substantive rules, clear evidentiary rules, and an opportunity to appeal. The forum needn’t be a public courtroom. Take, for example, the Copyright Alert System, which was implemented post-SOPA to adjudicate P2P file-sharing claims arising from residential broadband use. For all of CAS’s shortcomings, its architects made a real effort to ensure a fair and predictable alternative dispute resolution process for accused infringers. The system employed real arbitrators who (ostensibly) applied settled rules of copyright law. I’m not sure what the current status of that program is, because its overseers haven’t done any public reporting about it for years. But it had better bones by far than what the MPAA and its partner DNS registries rolled out last year.

So, I will celebrate the anniversary of the defeat of SOPA and PIPA today, because I think they were bad bills that grew out of an evidence-free, industry-captured policymaking process. But I won’t get all carried away about how we saved the Internet back in 2012, because much of what SOPA and PIPA would have accomplished has since come about by less transparent means and with even fewer procedural protections for accused wrongdoers. Five years is a century in the life of the Internet. In 2017, the campaign for friction-free online content removal goes far beyond the MPAA and copyrights. Now, the targets are amorphously identified fake news, hate speech, and terrorist content, in addition to piracy and counterfeiting. The roster of powerful complainants is expanding. And the largest intermediaries (Facebook, Microsoft, YouTube, Twitter) are putting up less resistance to categorical takedown demands than they did in the past. The Open Internet still needs saving, and the forces that threaten it now are both more complex and harder to contain than they were in 2012.