"Google's Gatekeepers" raises concerns and questions

There's a lot to say about Jeffrey Rosen's piece on "Google's Gatekeepers" in the Times Magazine on Sunday. Rosen shines much-needed light on the internal process for making decisions about removing content and responding to user and government requests for censorship at Google. Many have commented on the risks of entrusting so few people at a relatively small number of companies with so much power to determine what we see and don't see on the Internet, so I'll limit my comments to three areas of concern:

The role of government: Rosen pushes Michelle Wong, deputy general counsel and chief "decider" on content removal issues at Google, on what her ideal model for controlling content would look like. Wong and her colleagues respond by saying that

"they would be happiest, of course, if more countries would adopt U.S.-style free-speech protections. Knowing that that is unlikely, they said they would prefer that countries around the world set up accountable bodies that provide direct guidance about what controversial content to restrict."

In an ideal world, yes, governments would act in the interests of free speech and protecting human rights. But the reality of many governments around the world is a trend toward increased repression, decreased freedom and the use of the Internet to identify dissidents who express their views via email, VOIP, instant messaging, and social networking sites. Framing the debate as a choice between a system of ad hoc decision-making by a few individuals in companies on the one hand and decisions by democratic governmental institutions that will regulate content while protecting fundamental freedoms on the other, ignores the vast reality that exists between these two poles.

It's in this middle ground where new initiatives like the multi-stakeholder Global Network Initiative (GNI) and the proposed Global Online Freedom Act seek to increase accountability, transparency and reporting by companies. These new efforts are aimed at moving us closer to a global system that protects free speech to the greatest extent possible.

The development of the newly announced GNI is at a crucial moment. The next step of the initiative will focus on developing concrete benchmarks for company compliance with GNI principles and a governance structure for decision-making among the group's members. The measure of GNI's success will be whether it moves the industry toward greater transparency and accountability in the kinds of situations Rosen describes in this article.

Benevolent users: Rosen quotes Andrew McLaughlin, Google's director of global public policy, as expressing confidence in the power of global networks of millions of users "to agitate against censorship as they experienc[e] the benefits of free speech." Rosen cites evidence from Turkey, where bloggers have rallied against censorship efforts, as supporting McLaughlin's vision of widespread counter-censorship movements among users.

But not all users have the same perspective on free speech on the Internet or the political space in their own societies to make these challenges without serious retaliation for doing so. There's a new study out from the National Bureau of Economic Research, "Regulation and Distrust," which finds that people living under corrupt governments tend to want increased regulation. The study suggests a cycle of corruption, distrust and demand for more regulation from government, even when the government is widely viewed as corrupt.

Along the same lines, Deborah Fallows at the Pew Research Center published a study in 2007 which found that Chinese citizens actually desire more censorship of the Internet. The study argues that Chinese users distrust much of the information available on the Internet and also fear societal degradation through unregulated Internet access to things like child pornography.

So while examples like Turkey are heartening, they don't tell the whole story. We can't rely on a vision of benevolent governments and benevolent users to get us to a place where companies respect fundamental freedoms in a transparent and accountable way.

Decision-making processes: Throughout the piece, Rosen highlights the extremely informal nature of decision-making within Google. The process seems ultimately to be based on the "best guess" of Wong and her colleagues in each situation as it arises. To be sure, Wong, McLaughlin and Kent Walker, Google's general counsel, appear in Rosen's portrayal as qualified, competent and interested in expanding freedoms on the Internet.

But we have to wonder about those facts we don't know--Google granted Rosen significant access and information about their process for addressing censorship issues, but the company controlled what Rosen saw and heard, especially from Google employees. Entrusting a small group of people within a company to make the right decisions without mechanisms for transparency, reporting and accountability to other stakeholders simply does not go far enough.

Rosen gives us a brief glimpse into Yahoo! through an excerpt from an interview with Michael Samway, deputy general counsel and head of the company's Business & Human Rights Program (full disclosure: I was an intern in this program from June to August this year). Yahoo! has implemented a system to address challenges to human rights and as Rosen points out in the article, is adjusting its business model to locate communications data and search queries outside of repressive countries to limit their jurisdiction over sensitive information.

Though the company is struggling financially right now, its Business & Human Rights Program provides an interesting and ambitious model for a more predictable and structured approach to censorship and freedom of expression issues.

Rosen's piece is valuable because it gives users insight into what decision-making looks like at the biggest of the Internet companies. We should be looking for more information like this, and pushing for greater transparency on issues of censorship and freedom of expression not only at Google, but at Yahoo!, Microsoft, Cisco, Verizon, Comcast, Vodafone, French Telecom and other companies that influence control over information and data on the Internet.

The Internet and Human Rights

The Internet and human rights may not be immediately connected in our minds when we first think of email, blogging, shopping, YouTube, etc. But the Internet has a very real impact on the exercise of the rights to freedom of expression and privacy around the world. While the Internet has created new paths for dissidents, activists, journalists and individuals to share ideas and critique government action, it has come with a new set of concerns about censorship, surveillance and privacy.

Freedom of expression and privacy rights are protected in the International Covenant on Civil and Political Rights, an international human rights convention to which almost all countries are party. The covenant protects the right to hold opinions without interference, the right to freedom of expression (including the right to seek, receive and impart information and ideas in any form of media), and the right to privacy.

The Internet has opened up an entirely new—and often anonymous—way to express opinions and exchange information. When a government doesn’t care for particular viewpoints or ideas found on the Internet, it may try to curtail their expression through censorship and surveillance of online activity, and through intimidation and prosecution of the authors of the objectionable content.

A caveat: there’s a difference between sharing child porn on the Internet and sharing ideas about current events and basic freedoms. Where exactly that line falls is the subject of a different debate, but let me be clear that I’m not talking (for the moment) about legally authorized surveillance of Internet activity in pursuit of people breaking the law.

A few examples to highlight the intersection between human rights and the Internet:

The New York Times ran a story this week on the discovery that text messages sent using Tom-Skype, a joint venture between eBay (Skype’s parent company) and a Chinese company, are subject to surveillance and storage on Chinese servers. Nart Villeneuve, a researcher at the University of Toronto, uncovered the surveillance and identified a set of words that trigger storage of messages and the IP addresses of the computers used to send them. The list of words predictably includes “Taiwan” and “Olympics,” but also “milk powder,” reflecting concern about criticism of the government’s response to the discovery of tainted milk that has caused the deaths of at least four children and sickness in 50,000 others.

Secondly, Reporters Without Borders reports this week that a Malaysian blogger, Raja Petra Kamarudin is being tried for sedition in association with an article he published in the online newspaper Malaysia Today implicating members of the ruling party in the 2006 murder of a Mongolian woman.

These examples highlight the different ways surveillance and censorship take shape in countries where the government doesn’t tolerate the free expression of ideas, and particularly ideas that are critical of the government. Surveillance can be wholesale and general, targeting every Tom-Skype user in China, or it can be highly specific, targeting a particular journalist.

Whether it's protecting user data or providing Internet hardware to repressive governments, companies have an important role to play in respecting freedom of expression and privacy rights around the world. I’ll be following the discussion around the Internet and human rights, and particularly a new set of Internet principles that is expected to be announced soon addressing freedom of expression and privacy rights. Stay tuned.