This piece is exerpted from the Law, Borders, and Speech Conference Proceedings Volume. The conference, convened by Stanford's Center for Internet and Society, brought together experts from around the world to discuss conflicting national laws governing online speech -- and how courts, Internet platforms, and public interest advocates should respond to increasing demands for these laws to be enforced on the global Internet. For two weeks in January 2018, we will be posting these materials on the CIS Blog. The Proceedings Volume itself contains these and other resources, including reading lists, conference slides, and results of participant surveys. It is Creative Commons licensed for re-use in teaching materials and elsewhere.
Panel Summary by Graham Smith
- David Drummond - Senior Vice President, Corporate Development, Alphabet
- Mike Godwin - Senior Fellow, R Street Institute
- Joseph Lorenzo Hall - Chief Technologist, Center for Democracy & Technology
- Graham Smith - Partner, Bird & Bird, LLP
- Marketa Trimble - Samuel S. Lionel Professor of Intellectual Property Law, William S. Boyd School of Law, University of Nevada, Las Vegas
Technical tools can block Internet users from seeing certain content in their countries. How well do they work, what unintended consequences might they have, and is it a good idea for law to pressure private companies to adopt them?
David Drummond introduced the panel. The significance of this panel is that geoblocking tools are in many ways defining and enforcing jurisdiction.
A number of themes emerged from the panel.
What is geoblocking?
The session started with a divergence of definitions. Joseph Lorenzo Hall's technical introduction ranged widely over website and network blocking technologies and described some circumvention tools. Graham Smith distinguished between technical methods used by sites to refuse incoming requests (geoblocking) and techniques to prevent outbound requests reaching their destination (network blocking). Both were discussed extensively during the panel and discussion.
Good, bad or neither?
Marketa Trimble identified two purposes for which tools such as VPN and TOR could be used to circumvent geoblocking: (1) the need to be 'anywhere but' a particular country, which would be the concern of free speech activists and (2) to appear to be in a specific place, in order to watch a TV show available only in that location. Geoblocking and VPN should be regarded as neutral tools and the law should be qualified so that people don’t have to worry about negative legal implications of using them.
Although, as Trimble observed, geoblocking has been vilified, panelists noted some positive aspects. For example, Hall has blocked Chinese IP addresses from his josephhall.org website because users and search engines from the country were using up so much bandwidth. Mike Godwin noted that geoblocking could be used as a form of protest, such as when some sites blocked requests from the USA during the SOPA/PIPA ‘going dark' protest. Trimble suggested that there are situations and places where geoblocking has legitimate purposes and can be likened to someone placing a lock on their own house when they leave. The same lock on a polling station at election time would be problematic. It is a tool that can be misused like any other technical tool. Smith suggested that issues arise when geoblocking is over-incentivised or compelled.
Geoblocking or targeting?
Drummond stressed that geoblocking is often implemented reluctantly, as a less damaging alternative to global removal of content. He questioned why a targeting regime, in which users default to content on local domains but can still see content elsewhere if they take steps to do so, is not sufficient. This preserves the users' right to travel in cyberspace while for the most part ensuring that people see only local content, since the defaults are very powerful. This sort of regime handles the territorial problem quite well. Since people can get around geoblocking you are not getting a lot more law enforcement that way.
The right to travel in cyberspace
The idea of the right to travel in cyberspace provoked discussion. Trimble referred to her 2012 paper in which she had explored right to travel arguments. In the physical world there are limitations, such as needing a passport to travel abroad. In cyberspace, it is up for debate what kinds of limitations are appropriate in which situations. Drummond said that Google had always resisted geoblocking because while it did not deprive the whole world from seeing content in places where it is legal, it does eliminate the right to travel.
Smith observed that if your starting point is that your citizens, whether they were inside or outside your borders, should under no circumstances be able to access content governed by a different legal regime, then you are unlikely to think that a targeting regime works well. However, this approach imposes more restrictive borders than in the offline world. It is something more like the Berlin Wall, whose purpose was to keep people in. In cyberspace, keeping information out keeps people in.
Some discussion centered around granularity. Hall argued against establishing default rules that would effectively require constant, granular tracking of Internet and mobile users’ location, undermining their privacy. Godwin commented that we do not know what 100% effective blocking of objectionable content looks like except to totally Balkanize the Internet. We are now seeing both of the two worlds envisaged by Johnson/Post: lowest common denominator of content enforcement and Balkanization based on terrestrial jurisdictions.
Trimble commented that a digital passport would need, at a certain level of abstraction, information about where the user is from and what kind of content is available to it.
A prohibition arms race?
A recurring theme in the discussion was the tendency towards a prohibition arms race: blocking followed by circumvention followed by action against circumvention. At each stage prohibitions tend to become more general, leading to increasing negative effects as legitimate behaviour is affected. Hall noted that the use of VPN tunnels can be countered by a service blocking VPN exit points, which prevents those who need to use VPN from using the service. However, one new tool—called Streisand—runs multiple flavors of VPN, allowing increasing agility to users seeking to circumvent censorship. For example, the Streisand tool can make all communications look like credit card transactions, which censors are loathe to block, lest they impact Internet e-commerce in their region. As Hall and others have seen emerging in the field, governments may now be deploying machine learning techniques—effectively artificially-intelligent systems that can “learn” to spot leaks in the Great Firewall—to counter these new tools.
Trimble asked whether circumvention of geoblocking is a widespread practice leading to significant erosion of territoriality or had negligible spillover. Should we abandon attempts to achieve territoriality, or regulate the spillover?
Smith commented that Johnson/Post had predicted that attempts to preserve borders in cyberspace would prove futile. However, we have to bear in mind what timespan we are considering. We don't know if we are witnessing the brink of a new era of a Balkanized Internet or the last thrashings of the dinosaurs.
Big company rules for small actors?
The discussion moved on to the effect on small actors of making laws on the basis of what the large players can do, such as geoblocking. Will the local online newspaper simply block the rest of the world to avoid the problem of understanding the laws of every other country? What other unintended consequences may arise from geoblocking becoming a norm, if indeed it is a norm?
Godwin observed that Google has the resources to be responsive and make discriminations among cases. The chances are a new market entrant would not be able to do that. This raises the barrier to entry. Trimble emphasized the importance of the possibility of choice. A small newspaper told that it could not geoblock might have no option but to obtain a global or regional copyright license, which it might not be able to afford.
It was suggested that countries might adopt a twin strategy of controlling large companies providing protocols such as VPN through licensing requirements, and putting small disruptive technologists out of business. Hall said it would be unfortunate if pressure put on such developers led to them having to remain anonymous in order to develop secure communications tools.
The possibility of crafting different laws based on different sizes of company was also raised. That would require consideration of what is meant by size, when a small company is capable of having a large effect on the Internet.