My project was set up as an experiment to test our awareness and expectations of online privacy. My hypothesis was that Internet users are largely unaware of the amount of information accumulated and aggregated about them online. We have an “anonymity of Manhattan” mindset: we believe that no one will pick us out from the crowd of billions of users. Everyone publishes their name, email, birthday, and address for sign-ups. Everyone has incriminating pictures on Facebook, follows the Jersey Shore cast on Twitter, buys embarrassing books on Amazon, and looks up basic concepts on Wikipedia. We seem to think that it is a slim chance that someone, one day, would use our information against us. And even if they did, the cost of this chance does not outweigh the multitude of benefits the Internet gives us. For this reason, I predicted that while most people hope for some level of privacy online, they are largely apathetic to understanding and enforcing it.
To test that hypothesis, I focused on the privacy policies of five websites: Gmail, Facebook, Twitter, Amazon, and Wikipedia. I chose these sites for two reasons. First, they are all, of course, in the top 10 most visited sites globally. Second, searching myself on a site called PeekYou, which helps you find people online, I discovered a profile of myself that included links to my Google+, Facebook, Twitter, and Wikipedia accounts, as well as a list of my most recent Amazon purchases (and for $39.99 I could order a full background check on myself). Bewildered by how this information had been leaked on to PeekYou (and what could possibly be in my background check), I decided to focus on the policies of these five sites—I found these policies to be overwhelming similar.
I then wrote a survey on the website SurveyMonkey of 29 yes/no, fill-in, and multiple-choice questions, designed to be easily and quickly completed. Questions were divided to address four issues: demographics, knowledge and awareness, descriptive activity, and normative opinions. 173 responses were collected and I then underwent simple statistical analysis on Excel to identify possible correlations as well as discrepancies with the relevant privacy policies.
The results confirmed that we have very different assumptions about websites than their privacy policies stipulate and, overall the project provides a basis to begin answering the broader question of how we can engineer norms, laws, and code to make the architecture of the Internet fit our descriptive and normative desires, and not the other way around.
Happy holidays! -Carla
Yale Law Tech 