Complex, data-driven computer systems are a core part of modern life. Users often must make nuanced decisions when interacting with complicated systems and algorithms, yet they are rarely equipped to do so. This talk will discuss two streams of work that improve such interactions by co-designing better underlying models and user interfaces. First, I will discuss our work using formal models of an Internet of Things system to improve trigger-action programming (TAP), a popular end-user-programming method that lets users express how devices and services should interact. I will focus on how our TAPdiff system leverages our formal model to highlight differences in properties and outcomes across related variants of TAP rules. Second, I will present a series of projects that improve consumers’ understanding of ad-related online tracking and its consequences. I will highlight how having consumers request copies of their own Twitter data under recent privacy laws enabled us to illuminate the current ecosystem of ad targeting and improve privacy transparency mechanisms. I will conclude with our ongoing work aiming to make data-access rights more useful and informative for consumers.
Blase Ur is an Assistant Professor of Computer Science at the University of Chicago, where he researches security, privacy, human-computer interaction, and ethical AI. He is part of the UChicago SUPERgroup, which uses data-driven methods to help users make better security and privacy decisions, as well as to improve the usability of complex computer systems. He has received an NSF CAREER Award (2021), three best paper awards, five honorable mention paper awards, and UChicago’s Quantrell Award for undergraduate teaching (2021). He holds degrees from Carnegie Mellon University (PhD and MS) and Harvard University (AB).