Research

My main interest for a long time has been helping people make sense of and manage information, both individually and as groups. More recently this has grown to include leveraging people's current behaviors online, along with social science theory, to produce individual and social goods that otherwise would not have been created.

More details on what this means are available in my research statement. For folks who want the overview of publications and collaborators, you can download my CV or my publications, or visit my Google Scholar profile.

Active Projects

These interests lead me into a number of cool domains; below are a few of the major ongoing themes. More details, and other projects, are available at the Reimagination Lab website.

Participation in (civic, deliberative) online communities

I've been working on understanding and designing for participation in online communities for a long time, including MovieLens and Wikipedia (where I wrote the original version of SuggestBot). I tend to think about designs that map well to psychological and sociological theories about motivations for reading, sharing, and contributing content, with the hope that developing these designs provides technical novelty, theoretical insight, and practical benefit.

These days, my main involvement in this topic is through a research stream led by Brian McInnis, who I co-advise along with Gilly Leshed. Brian's done a rich set of studies around how crowd workers might contribute to large-scale deliberative discussions, tightly marrying ideas from the crowdwork literature with the theory and practice of facilitating deliberative discourse. This has led to interesting contributions about modeling and supporting newcomers' ability to productively contribute to deliberations. It also helped him develop policy and design recommendations around the design of Amazon Mechanical Turk based on participants' discussion of how Turk's policies around rejecting work place workers at risk.

Benefits and risks of self-disclosure online

People regularly disclose information about themselves online that is regularly used by both companies and researchers to make inferences about people's preferences, attitudes, and future behaviors. Much of my career arc has looked to leverage these data to improve people's experience of recommender systems and social media (including some of the work described above to support effective participation). Individual people also use these disclosures, both for forming impressions about and making decisions whether to interact with others, and (as in my early Pensieve work and the commercial systems Timehop and Facebook's Memories feature) as a personal archive that can support reminiscing, reflection, and self-understanding. This power of disclosed personal data raises natural questions about what can and should be done with it, and the benefits and risks of self-disclosure.

These days I am collaborating with folks on two main lines of related work. The first, led by Shruti Sannon, focuses on how people proactively manage their privacy concerns when asked for personal information by both companies and by other individuals. Shruti's come up with a number of contributions about how privacy concerns translate into privacy protective behaviors in different contexts -- specifically, privacy lies told to protect private information -- and the value of thinking about privacy through the lens of deception and morality. Through conducting the study on Mechanical Turk, Shruti found a number of stories of how the power relationships between employers and workers on Turk influence privacy decisions, leading to a study that will be published at CHI 2019.

The second, led by Yoon Hyung Choi and Natalie Bazarova, looks at how the combination of people's mental state and their perceptions of a communication channel's properties affects their concerns about self-presentation, their decisions to disclose personal information, the style and content of those disclosures, the way people react to those disclosures, and the social and psychological benefits people experience through those disclosures and responses.

Network-centric recommendation and sharing in social networks

This project, now quiescent but still worth listing, started from a question Amit Sharma asked: What if recommender systems were designed, not for individuals and purchasing, but from the ground up for social networks? What kinds of questions, use cases, metrics, and algorithms would emerge, and how would they be different from traditional recommender systems research?

One set of questions we've focused on is at the micro-level of sharing between individuals: how do people choose to share particular items with particular audiences, and what makes people accept the things other people share? Social influence, homophily, trust, and personal preferences all likely affect these decisions, making it important to account for them in models of sharing behavior and systems that support it.

A second set is at the macro level of how items diffuse in social networks: what effect do those micro-level choices have on the patterns of preferences and diffusion we observe in real networks, and are there regularities between networks? We'd like to build nuanced diffusion models that account for these micro-level choices and explain the diffusion observed in networks better than current models.

Potential students

Note that I will be away from Cornell until after summer 2019, and possibly again after spring 2020. Thus, I'm unlikely to take on new folks right now. That said...

Motivated, passionate students of any level who like this work and want to get involved are always fun to talk to. Drop me an e-mail that makes clear why we're a good fit, why you're interested, what your goals and ideas are, and what you want to contribute. It won't always work out--at times I have more or less need, funding, and advising energy--but I'm generally happy to chat.

Teaching

I care deeply about teaching both undergraduate and graduate students. Most students and observing professors regard me as a solid classroom teacher, and I try to be a supportive, flexible advisor. I try to give students space to take assignments and projects in directions of their own interest, plenty of hands-on work both as individuals and groups, and copious support.

For evidence and some philosophy, see my teaching statement and advising statement.

Current courses

I am away from Cornell and thus not teaching courses in calendar year 2016.

Past courses

A typical teaching scene from the 2012 version of HCI (pic by Chelsea Howe).