May 25, 2018

(Mis)conceptions About the Impact of Surveillance

Does surveillance impact behavior? Or is its effect, if real, only temporary or trivial? Government surveillance is back in the news thanks to the so-called “Nunes memo”, making this is a perfect time to examine new research on the impact of surveillance. This includes my own recent work, as my doctoral research at the Oxford Internet Institute, University of Oxford  examined “chilling effects” online, that is, how online surveillance, and other regulatory activities, may impact, chill, or deter people’s activities online.

Though the controversy surrounding the Nunes memo critiquing FBI surveillance under the Foreign Intelligence Surveillance Act (FISA) is primarily political, it takes place against the backdrop of the wider debate about Congressional reauthorization of FISA’s Section 702, which allows the U.S. Government to intercept and collect emails, phone records, and other communications of foreigners residing abroad, without a warrant. On that count, civil society groups have expressed concerns about the impact of government surveillance like that available under FISA, including “chilling effects” on rights and freedoms. Indeed, civil liberties and rights activists have long argued, and surveillance experts like David Lyon long explained, that surveillance and similar threats can have these corrosive impacts.

Yet, skepticism about such claims is common and persistent. As Kaminski and Witov recently noted, many “evince skepticism over the effects of surveillance” with deep disagreements over the “effects of surveillance” on “intellectual queries” and “development”.  But why?  The answer is complicated but likely lies in the present (thin) state of research on these issues, but also common conceptions, and misconceptions, about surveillance and impact on people and broader society.

Skepticism and assumptions about impact
Skepticism about surveillance impacts like chilling effects is, as noted, persistent with commentators like Stanford Law’s Michael Sklansky insisting there is “little empirical support” for chilling effects associated with surveillance or Leslie Kendrick, of UVA Law, labeling the evidence supporting such claims “flimsy” and calling for more systematic research on point. Part of the problem is precisely this: the impact of surveillance—both mass and targeted forms—is difficult to document, measure, and explore, especially chilling effects or self-censorship. This is because demonstrating self-censorship or chill requires showing a counterfactual state of affairs: that a person would have said something or done something but for some surveillance threat or awareness.

But another challenge, just as important to address, concerns common assumptions and perceptions as to what surveillance impact or chilling effects might look like. Here, both members of the general public as well as experts, judges, and lawyers often assume or expect surveillance to have obvious, apparent, and pervasive impact on our most fundamental democratic rights and freedoms—like clear suppression of political speech or the right to peaceful assembly.

A great example of this assumption, leading to skepticism about whether surveillance may promote self-censorship or have broader societal chilling effects—is here expressed by University of Chicago Law’s Eric Posner. Posner, a leading legal scholar who also incorporates empirical methods in his work, conveys his skepticism about the “threat” posed by National Security Agency (NSA) surveillance in a New York Times “Room for Debate”  discussion, writing:

This brings me to another valuable point you made, which is that when people believe that the government exercises surveillance, they become reluctant to exercise democratic freedoms. This is a textbook objection to surveillance, I agree, but it also is another objection that I would place under “theoretical” rather than real.  Is there any evidence that over the 12 years, during the flowering of the so-called surveillance state, Americans have become less politically active? More worried about government suppression of dissent? Less willing to listen to opposing voices? All the evidence points in the opposite direction… It is hard to think of another period so full of robust political debate since the late 1960s—another era of government surveillance.

For Posner, the mere existence of “robust” political debate and activities in society is compelling evidence against claims about surveillance chill.

Similarly, Sklansky argues not only that there is “little empirical support” for the claim that surveillance would “chill independent thought, robust debate, personal growth, and intimate friendship”— what he terms “the stultification thesis”—but like Posner, he finds persuasive evidence against the claim “all around us”. He cites, for example, the widespread “sharing of personal information” online (which presumably would not happen if surveillance was having a dampening effect); how employer monitoring has not deterred employee emailing nor freedom of information laws deterred “intra-governmental communications”; and how young people, the “digital natives” that have grown up with the internet, social media, and surveillance, are far from stultified and conforming but arguably even more personally expressive and experimental than previous generations.  In light of all that, Sklansky dismisses surveillance chill as simply not “worth worrying about”.

I sometimes call this the “Orwell effect”—the common assumption, likely thanks to the immense impact Orwell’s classic novel 1984 has had on popular culture, that surveillance will have dystopian societal impact, with widespread suppression of personal sharing, expression, and political dissent. When Posner and Sklansky (and others that share these common expectations) do not see these more obvious and far reaching impacts, they then discount more subtle and less apparent impacts and effects that may, over the long term, be just as concerning for democratic rights and freedoms. Of course, theorists and scholars like Daniel Solove have long interrogated and critiqued Orwell’s impact on our understanding of privacy and Sklansky is himself wary of Orwell’s influence, so it is no surprise his work also shapes common beliefs and conceptions about the impact of surveillance.  That influence is compounded by the earlier noted lack of systematic empirical research providing more grounded insights and understanding.

This is not only an academic issue. Government surveillance powers and practices are often justified with reference to other national security concerns and threats like terrorism, as this House brief on the FISA re-authorization illustrates. If concerns about chilling effects associated with surveillance and other negative impacts are minimized or discounted based on misconceptions or thin empirical grounding, then challenging surveillance powers and their expansion is much more difficult, with real concrete implications for rights and freedoms.

So, the challenge for documenting, exploring, and understanding the impact of surveillance is really two-fold. The first is one of research methodology and design: designing research to document the impact of surveillance, and a second concerns common assumptions and perceptions as to what surveillance chilling effects might look like—with even experts like Posner or Sklansky assuming widespread speech suppression and conformity due to surveillance.

New research, new insights
Today, new systematic empirical research on the impact of surveillance is being done, with several recent studies having documented surveillance chilling effects in different contexts, including recent studies by  Stoycheff [1], Marthews and Tucker [2], as well as my own recent research.  This includes an empirical legal study[3] on how the Snowden revelations about NSA surveillance impacted Wikipedia use—which received extensive media coverage in the U.S. and internationally— and a more recent study[4], which I wrote about recently in Slate, that examined among other things how state and corporate surveillance impact or “chill” certain people or groups differently. A lot of this new work was not possible in previous times, as it is based on new forms of data being made available to researchers and insights gleaned from analyzing public leaks and disclosures concerning surveillance like the Snowden revelations.

The story these and other new studies tell when it comes to the impact of surveillance is more complicated and subtle, suggesting the common assumptions of Posner and Sklansky are actually misconceptions. Though more subtle, these impacts are no less concerning and corrosive to democratic rights and freedoms, a point consistent with the work of surveillance studies theorists like David Lyon[5] and warnings from researchers at places like the Citizen Lab[6], Berkman Klein Center[7], and here at the CITP[8].  In subsequent posts, I will discuss these studies more fully, to paint a broader picture of surveillance effects today and, in light of increasingly sophisticated targeting and emerging automation technologies, tomorrow. Stay tuned.

* Jonathon Penney is a Research Affiliate of Princeton’s CITP, a Research Fellow at the Citizen Lab, located at the University of Toronto’s Munk School of Global Affairs, and teaches law as an Assistant Professor at Dalhousie University. He is also a research collaborator with Civil Servant at the MIT Media Lab. Find him on twitter at @jon_penney

[1] Stoycheff, E. (2016). Under Surveillance: Examining Facebook’s Spiral of Silence Effects in the Wake of NSA Internet Monitoring. Journalism & Mass Communication Quarterly. doi: 10.1177/1077699016630255

[2] Marthews, A., & Tucker, C. (2014). Government Surveillance and Internet Search Behavior. MIT Sloane Working Paper No. 14380.

[3] Penney, J. (2016). Chilling Effects: Online Surveillance and Wikipedia Use. Berkeley Tech. L.J., 31, 117-182.

[4] Penney, J. (2017). Internet surveillance, regulation, and chilling effects online: A comparative case study. Internet Policy Review, forthcoming

[5] See for example: Lyon, D. (2015). Surveillance After Snowden. Cambridge, MA: Polity Press; Lyon, D. (2006). Theorizing surveillance: The panopticon and beyond. Cullompton, Devon: Willan Publishing; Lyon, D. (2003). Surveillance After September 11. Cambridge, MA: Polity. See also Marx, G.T., (2002). What’s New About the ‘New Surveillance’? Classifying for Change and Continuity. Surveillance & Society, 1(1), pp. 9-29;  Graham, S. & D. Wood. (2003). Digitising Surveillance: Categorisation, Space, Inequality, Critical Social Policy, 23(2): 227-248.

[6] See for example, recent works: Parsons, C., Israel, T., Deibert, R., Gill, L., and Robinson, B. (2018). Citizen Lab and CIPPIC Release Analysis of the Communications Security Establishment Act. Citizen Lab Research Brief No. 104, January 2018; Parsons, C. (2015). Beyond Privacy: Articulating the Broader Harms of Pervasive Mass Surveillance. Media and Communication, 3(3), 1-11; Deibert, R. (2015). The Geopolitics of Cyberspace After Snowden. Current History, (114) 768 (2015): 9-15; Deibert, R. (2013) Black Code: Inside the Battle for Cyberspace, (Toronto: McClelland & Stewart).  See also

[7] See for example, recent work on the Surveillance Project, Berkman Klein Center for Internet and Society, Harvard University.

[8] See for example, recent work: Su, J., Shukla, A., Goel, S., Narayanan, A., De-anonymizing Web Browsing Data with Social Networks. World Wide Web Conference 2017; Zeide, E. (2017). The Structural Consequences of Big Data-Driven Education. Big Data. June 2017, 5(2): 164-172, https://doi.org/10.1089/big.2016.0061;MacKinnon, R. (2012) Consent of the networked: The worldwide struggle for Internet freedomNew YorkBasic Books.; Narayanan, A. & Shmatikov, V. (2009). See also multiple previous Freedom to Tinker posts discussing research/issues point.

 

How Will Consumers Use Faster Internet Speeds?

This week saw an exciting announcement about the experimental deployment of DOCSIS 3.1 in limited markets in the United States, including Philadelphia, Atlanta, and parts of northern California, which will bring gigabit-per-second Internet speeds to many homes over the existing cable infrastructure. The potential for gigabit speeds over the existing cable networks bring hope that more consumers will ultimately enjoy much higher-speed Internet connectivity both in the United States and elsewhere.

This development is also a pointed response to the not-so-implicit pressure from the Federal Communications Commission to deploy higher-speed Internet connectivity, which includes other developments such as the redefinition of broadband to a downstream throughput rate of 25 megabits per second, up from a previous (and somewhat laughable) definition of 4 Mbps; many commissioners have also stated their intentions to raise the threshold for the definition of a broadband network to a downstream throughput of 100 Mbps, as a further indication that ISPs will see increasing pressure for higher speed links to home networks. Yet, the National Cable and Telecommunications Association has also claimed in an FCC filing that such speeds are far more than a “typical” broadband user would require.

These developments and posturing beg the question: How will consumers change their behavior in response to faster downstream throughput from their Internet service providers? 

Ph.D. student Sarthak Grover, postdoc Roya Ensafi, and I set out to study this question with a cohort of about 6,000 Comcast subscribers in Salt Lake City, Utah, from October through December 2014. The study involved what is called a randomized controlled trial, an experimental method commonly used in scientific experiments where a group of users is randomly divided into a control group (whose user experience no change in conditions) and a treatment group (whose users are subject to a change in conditions).  Assuming the cohort is large enough and represents a cross-section of the demographic of interest, and that the users for the treatment group are selected at random, it is possible to observe differences between the two groups’ outcomes and conclude how the treatment affects the outcome.

In the case of this specific study, the control group consisted of about 5,000 Comcast subscribers who were paying for (and receiving) 105 Mbps downstream throughput; the treatment group, on the other hand, comprised about 1,500 Comcast subscribers who were paying for 105 Mbps but at the beginning of the study period were silently upgraded to 250 Mbps. In other words, users in the treatment group were receiving faster Internet service but was unaware of the faster downstream throughput of their connections. We explored how this treatment affected user behavior and made a few surprising discoveries:

“Moderate” users tend to adjust their behavior more than the “heavy” users. We expected that subscribers who downloaded the most data in the 250 Mbps service tier would be the ones causing the largest difference in mean demand between the two groups of users (previous studies have observed this phenomenon, and we do observe this behavior for the most aggressive users). To our surprise, however, the median subscribers in the two groups exhibited much more significant differences in traffic demand, particularly at peak times.  Notably, the 40% of subscribers with lowest peak demands more than double their daily peak traffic demand in response to service-tier upgrades (i.e., in the treatment group).

With the exception of the most aggressive peak-time subscribers, the subscribers who are below the 40th percentile in terms of peak demands increase their peak demand more than users who initially had higher peak demands.

This result suggests a surprising trend: it’s not the aggressive data hogs who account for most of the increased use in response to faster speeds, but rather the “typical” Internet user, who tends to use the Internet more as a result of the faster speeds. Our dataset does not contain application information, so it is difficult to say what, exactly is responsible for the higher data usage of the median user. Yet, the result uncovers an oft-forgotten phenomena of faster links: even existing applications that do not need to “max out” the link capacity (e.g., Web browsing, and even most video streaming) can benefit from a higher capacity link, simply because they will see better performance overall (e.g., faster load times and more resilience to packet loss, particularly when multiple parallel connections are in use). It might just be that the typical user is using the Internet more with the faster connection simply because the experience is better, not because they’re interested in filling the link to capacity (at least not yet!).

Users may use faster speeds for shorter periods of time, not always during “prime time”. There has been much ado about prime-time video streaming usage, and we most certainly see those effects in our data. To our surprise, the average usage per subscriber during prime-time hours was roughly the same between the treatment and control groups, yet outside of prime time, the difference in usage was much more pronounced between the two groups, with average usage per subscriber in the treatment group exhibiting 25% more usage than that in the control group for non-prime-time weekday hours.  We also observe that the peak-to-mean ratios for usage in the treatment group are significantly higher than they are in the control group, indicating that users with faster speeds may periodically (and for short times) take advantage of the significantly higher speeds, even though they are not sustaining a high rate that exhausts the higher capacity.

These results are interesting for last-mile Internet service providers because they suggest that the speeds at the edge may not currently be the limiting factor for user traffic demand. Specifically, the changes in peak traffic outside of prime-time hours also suggest that even the (relatively) lower-speed connections (e.g., 105 Mbps) may be sufficient to satisfy the demands of users during prime-time hours. Of course, the constraints on prime-time demand (much of which is largely streaming) likely result from other factors, including both available content and perhaps the well-known phenomena of congestion in the middle of the network, rather than in the last mile. All of this points to the increasing importance of resolving the performance issues that we see as a result of interconnection. In the best case, faster Internet service moves the bottleneck from the last mile to elsewhere in the network (e.g., interconnection points, long-haul transit links); but, in reality, it seems that the bottlenecks are already there, and we should focus on mitigating those points of congestion.

Further reading and study. You’ll be able to read more about our study in the following paper: A Case Study of Traffic Demand Response to Broadband Service-Plan Upgrades. S. Grover, R. Ensafi, N. Feamster. Passive and Active Measurement Conference (PAM). Heraklion, Crete, Greece. March 2016. (We will post an update when the final paper is published in early 2016.) There is plenty of room for follow-up work, of course; notably, the data we had access to did not have information about application usage, and only reflected byte-level usage at fifteen-minute intervals. Future studies could (and should) continue to study the effects of higher-speed links by exploring how the usage of specific applications (e.g., streaming video, file sharing, Web browsing) changes in response to higher downstream throughput.

Where is Internet Congestion Occurring?

In my post last week, I explained how Netflix traffic was experiencing congestion along end-to-end paths to broadband Internet subscribers, and how the resulting congestion was slowing down traffic to many Internet destinations. Although Netflix and Comcast ultimately mitigated this particular congestion episode by connecting directly to one another in a contractual arrangement known as paid peering, several mysteries about the congestion in this episode and other congestion episodes that persist. In the congestion episodes between Netflix and Comcast in 2014, perhaps the biggest question concerns where the congestion was actually taking place. There are several theories about where congestion was occurring; one or more of them are likely the case. I’ll dissect these cases in a bit more detail, and then talk more generally about some of the difficulties with locating congestion in today’s Internet, and why there’s still work for us to do to shed more light on these mysteries.
[Read more…]