Lumen Researcher Interview Series: Professor Jon Penney

The Lumen team had the privilege earlier this month to reconnect with Professor Jon Penney, a long-time friend of Lumen, whose research has not only been critical to helping understand how internet users understand their interactions with the takedown ecology, but has also frequently made use of Lumen data. We took the oppportunity to catch up with Professor Penney at length, and to ask him about his experiences as a Lumen researcher.

The conversation, some of which took place over email, has been edited for length and clarity.

Professor, you wear a wide variety of professional hats. Could you summarize them for Lumen's audience?

I am currently an Associate Professor at Osgoode Hall Law School, York University. I am also a Research Fellow at the University of Toronto's Citizen Lab, a Visiting Scholar at the Institute for Rebooting Social Media, Harvard University, and a Faculty Associate at Harvard's Berkman Klein Center for Internet & Society. Before that, I was a long-time BKC community member, first as a Fellow, then a Research Affiliate.

It really does seem like Lumen has known you forever! On that topic, can you tell us about the papers, articles, or work in general where you’ve used Lumen in your research, and what led up to them?

There are four key papers here. In chronological order, they are:

First, my doctoral dissertation at Oxford, "Chilling Effects in the Internet Age: Three Case Studies” (DPhil Thesis, University of Oxford, November 2015 [unpublished]) (peer reviewed). I successfully defended it in November, 2015 with no corrections, and had it confirmed December 2016. My degree was officially awarded in May 2018.

In this work, I explored the phenomena of regulatory chilling effects online through three empirical legal case studies, one involving the impact of online surveillance, another on the impact of the digital copyright enforcement online, and another survey-based study comparing the impact of different forms of online state actions/regulation.

The Lumen database was critical to the empirical study of digital copyright enforcement online -- I examined a sample of Google Blogs and Twitter accounts that received DMCA takedown notices to explore the impact of the notices on the content targeted. My methodology was qualitative and quantitative content analysis.

Lumen provided me with the full collection of DMCA notices received over several, from which I drew randomized samples. I was able to locate relevant blogs/Twitter accounts using the notices themselves, to conduct the study

Next is “Privacy and Legal Automation: The DMCA as a Case Study” (2019) 22(1) Stanford Technology Law Review 412.
This is a law review article in which I published empirical findings from my doctoral thesis on the impact of DMCA notices sent to Google Blogs/Twitter accounts, and combined those findings with insights from a survey study, that explored hypothetical scenarios akin to internet users receiving DMCA notices (or similar personally legal threats). I feel comfortable saying that this publication would not be possible without Lumen.

After that are two papers I think should be mentioned as a pair.

“Internet Surveillance, Regulation, and Chilling Effects Online: A Comparative Case Study”, (2017) 6(2) Internet Policy Review 1 (peer reviewed)

and

“When Laws Free Us To Speak” (with Danielle Keats Citron) (2019) 87(6) Fordham Law Review 2317.

Both of these papers were informed by my earlier findings concerning DMCA notice impacts (made possible by Lumen). In other words, Lumen's database was critical to the conceptualization, analysis, and discussion of the studies (and findings) in these papers.

Finally, there is this one

Matias, J. N., Mou, M. E., Penney, J., & Klein, M. (2020). "Do Automated Legal Threats Reduce Freedom of Expression Online? Preliminary Results from a Natural Experiment." https://osf.io/nc7e2/

To explain this one, it's probably easiest to just share the abstract.

"Do copyright take-downs chill freedom of expression? To find out, we collected public tweets by 9,818 Twitter accounts that received a DMCA take-down notice in January and Febuary 2020. We then conducted an interrupted time-series analysis to study changes in people’s Twitter behavior after their content was taken down by Twitter. We found that, on average, those who received DMCA notices tweeted less frequently following the notice."

Again, the research and paper would not be possible without the Lumen database, which was critical to gathering and analysis the data.

I should also mention that I have a forthcoming book on chilling effects which also (not surprisingly) draws on Lumen supported research!

In all of these studies, Lumen made it possible to test hypothesis about the impact of DMCA notices specifically, and as a case study, automated legal enforcement online. The notices allowed us to track users and user accounts that had received DMCA notices. This led to critical insights as to the impact of online copyright enforcement, but since the DMCA is now often fully automated, it provides insight into the future of automated law as well.

Our methodologies included qualitative/quantitative content analysis coupled with quantitative statistical analysis, like interrupted time series analysis.Without the Lumen notices, we would simply not be able to conduct this research. There is no alternative.

How did you first learn about Lumen?

I first heard of Lumen/Chilling Effects database when I was a doctoral student at Oxford. Its existence influenced my doctoral work and later it would be among the key reasons I would apply for a Berkman Fellowship in 2012, when I was still a doctoral student at the Oxford Internet Institute -- because Lumen was so critical to my work, it made sense to become a member of the community. It's one of the best career decisions I've made, and I am proud to maintain ties to BKC today.

How often do you use Lumen in your research?

I still regularly consider Lumen in my projects and am still involved in a collaboration with the Citizens and Technology Lab (CAT Lab) based at Cornell, which explores automated legal enforcement via DMCA notices provided by Lumen.

From your perspective as a scholar, how important, in your opinion, is transparency through takedown notice sharing and related data?

As a scholar, I believe the Lumen database and the research and transparency it makes possible, is absolutely critical to fostering a safer, fairer, and more robust internet. There remains so much more work to be done, and strongly believe Lumen will be a key part of research and transparency for other emerging technologies, from AI and beyond.