A visit to the University of Michigan’s Interactive and Social Computing group
This past Tuesday and Wednesday I visited the University of Michigan’s Interactive and Social Computing Group (spanning faculty and students at the School of Information and Computer Science). I came give an invited talk about my lab’s work on effective, equitable, scalable learning of computing. But to me, visiting universities is about more than just disseminating my discoveries: it’s also about meeting with vibrant interdisciplinary communities to network, to learn new ideas, to recruit students, and sometimes to informally mentor.
The faculty and students at Michigan were diverse and incredibly sharp. I first met with a small group of doctoral students (Sang Won Lee, Harmanpreet Kaur, and Rebecca Krosnick) about a range of topics related to creative expression, microtask productivity, and programming skill. We talked a lot about the difference between fishing poles (developer tools) and fishing (programming skill) and whether tools can promote learning, and if so, what kind. We also talked about the nature of explicit strategies, and how strategies might mediate some of the phenomenon they were studying.
Next I met with Mark Newman, who has recently started investigating mobile computing and its connections to health and personal informatics. We had a fascinating conversation about some of the challenges of interdisciplinary research, and the untapped potential of deep investigations into other problem domains to surface insight into data, data curation, and software architectures. We also reflected and debated on what knowledge designers need about software engineering, if any, and what depth of technical understanding of software might be necessary to learn that knowledge. My book on Cooperative Software Development came up as one potential answer.
Next, I spoke. The MISC group has a nice model of having lunch 30 minutes before the seminar, so that everyone arrives early and on time for the talk, while allowing people to just show up for the talk. The audience was great, asking smart questions, probing skeptically into some of my results, while also connecting the ideas to the broad range of topics in learning analytics, social computing, and human computation. There was a fair number of questions about implications for teaching, but also about the implications for research on developer tools.
After the seminar, I met with another group of students (Brian Hall, Jason Brill, Ihudiya Finda Ogburu, Shiyan Yan, and other lurkers). This conversation was rich with mentoring, covering the importance of choosing a research community as a doctoral student, and the challenge of finding an advisor that can teach you community norms. We also talked about the inherent but important risks of doing interdisciplinary research that arise from the extra work bridging fields, importing and exporting theories to and from fields, and leaving one’s comfort zone to collaborate with people in other disciplines. I argued that such work isn’t for everyone, but can be very powerful.
Stephanie Teasley asked me a lot of questions about why I chose to focus on teaching the content of programming rather than trying to address the more acute sociocultural and inclusion challenges in computing. I admitted that while I feel those issues are more fundamental, as someone who designs interventions, I’m also less well-equipped to intervene on the cultures that cause these issues. We also discussed some of the bottom-up strategies that might promote better learning of programming skill, like promoting self-regulated learning through coaching, pair programming as a way of surfacing self-regulation, and the potential of learning analytics to mine these overt signals of self-regulation.
Next I met with Steven Oney, my academic sibling, as we had the same advisor. We talked about life as new faculty, his transition from his Presidential Scholar postdoc, and his fascinating new work on building computer-mediated communication for peer tutoring. His work has the exciting potential to scale the kinds of peer tutoring that happens informally in computer labs during CS classes, while also making this tutoring visible to systems.
Matt Kay, one of our recent University of Washington graduates, talked about his work on bias in data visualizations. He’s investigating the comprehensibility of frequentist versus Bayesian models and how data visualizations might mediate comprehensibility. I speculated that the learnability of visualizations might be another mediator, and that eco-centric human scale metrics (e.g., to offset your carbon footprint, you have to plant and care for a tree once every six hours) might be most learnable. Mapping this to data visualizations, the question is how easily someone can project their experiences into a data visualization.
Paul Resnick and I had a compelling conversation about a range of new projects he’s doing on learning, including spaced practice, prediction as pedagogy, and his collaboration with Steven Oney on peer tutoring. He also discussed the UMSI 106 course, which is an alternative to the intro CS course, and how they managed to achieve 50% female participation through an intentional design effort at inclusion.
My final meeting before dinner was with Tawanna Dillahunt. Tawanna was a new Ph.D. student when I was graduating, and she’s now already in her fourth year as as faculty. She’s doing truly novel work with underserved, high-poverty populations in Michigan, inventing interesting new systems for helping people describe their skills, find work opportunities, and plan transportation on poor public transit. We also chatted about some of the challenges of scaling this work, and the professional risk that come with deliberate, high impact dissemination.
Dinner was with Matt Kay, Walter Lasecki, Sarita Schoenebeck, and Mark Ackerman at Sava. As usual for dinners, we talked about lot more than research. I learned about Ann Arbor, it’s concentrated but notable gentrification, it’s surprisingly good food, and the even more surprising undeveloped land just miles outside the city. We also discussed many of the challenges and opportunities in scaling undergraduate research advising, the many unexpected experiences adolescents have with the internet, the policy, regulation, and ethics of information addition, the hidden complexities of measurement, academic study of conspiracy theories. We ended with one of the wackier ideas I’ve heard in a while: microtasks for micropeople (which I also described as mTykes or small human computation). No, this idea needs no further elaboration.
Because my flight left in the afternoon on Wednesday, I met with more folks in the morning. My first meeting was with Kentaro Toyama, who has been arguing for a theory of technology as an amplifier of social norms and structures, rather than a driver of social change. We probed into whether this is also true of technologies that promote learning, how to integrate these ideas into design education, and how to combine design education with instruction on information and design ethics. Kentaro challenged me to decide whether advancing the design of technology for any social goal is ever more important than directly advancing social change.
My last meeting was with professor Joyojeet Pal, who has worked on ICT4D and accessibility. We discussed some of the challenges motivating students to learn about accessibility, some of the cultural motivational differences between our institutions that lead to different levels of interest, and some of the visions from TeachAccess about needing both shallow and deep expertise on accessibility amongst future designers and engineers. We envisioned a role for integrating accessibility throughout curricula, but also the need for a deeper learning, perhaps in a Master’s in Accessible Technology to train the experts that industry desperately needs.
Finally, what better way to end a great visit to a great food town than a warm bowl of ramen? This was from Tomukun Noodle Bar, a lively place off campus for ramen, udon, and Korean BBQ. Thanks Ann Arbor, and thanks to everyone at the University of Michigan for a stimulating visit!
