Wicked Smart: 5 questions with U of M PhD and Wikipedian Aaron Halfaker



Aaron Halfaker

Five questions with Minnesota computer scientist (GroupLens PhD) and Wikimedia Foundation social researcher Aaron Halfaker.

Where are you from originally and how did you end up at GroupLens / The University of Minnesota?

I grew up in Virginia, MN, a mining town on the Iron Range. I was always interested in technology as a kid, but I didn’t always have access to it. When I started high school, my family had an old 386 that didn’t have any games on it, so I spent a lot of time making ASCII art in Word Star.

By the time I graduated from the Virginia High School, I had built my own computer and organized a couple of LAN parties.

I didn’t really see a future for myself in technology at the time. When I started at The College of St. Scholastica, I was a physical therapy major. I had a thing for science and medicine was/is a growing field. But in my first semester, I took an introductory programming class with Prof. Diana Johnson (who turned out to be an excellent instructor and mentor). I was hooked. Up until then, I thought programming as a mostly bland and dull process similar to doing algebra homework. Instead, I found that programming is a profoundly creative experience.

At the time, St. Scholastica was still developing it’s major and the classes in computer science were limited. I filled a substantial portion of the rest of my time at there with independent study and research. Due to some diligent prodding from my advisor, Prof. Jen Rosato (also excellent), I completed an application to the University of Minnesota’s graduate program. Eventually I found my thesis advisor, Prof. John Riedl, and the GroupLens Research lab.

What are you currently focused on?

My research occurs at an intersection of social science and information systems. I study crowds of people who use online communities to collaboratively build things and the technologies that make such mass collaboration possible. Wikipedia is where I’ve focused most of my efforts. My goal is to build a coherent theory around software systems can best support efforts like Wikipedia so that we can improve the system’s efficiency and replicate its success more broadly.

What motivates and inspires you do do this?

In 2007, the community of volunteer editors who do most of the work building and maintaining Wikipedia began to decline in size after 6 years of sustained growth. When I began looking for potential explanations of this abrupt switch, I began to start considering what the world would be like without Wikipedia — if this experiment in mass collaboration was doomed. In my opinion, Wikipedia and information resources like it the most important things for the future of humanity. When’s the last time you picked up a traditional print reference? When’s the last time you looked at a Wikipedia article? You’re not alone. According to the best estimates of unique readers, about 1/10th of the world reads Wikipedia at least once per month — and the fastest growing population of readers are from the developing world.

It’s not just Wikipedia. Stack Overflow has done a world of benefit for programming information and practical examples. Reddit has become influential enough to attract a US president, astronauts and entertainers to interact with its user base. User-generated content communities are becoming very influential. We ought to understand them so that we (humanity) can get the most out of them.

What impact is your research effort having?

My goal in my work with the Wikimedia Foundation is straightforward.  I want Wikipedia to be successful.  Right now, we have a very limited understanding of how collaborative, open information communities like Wikipedia work and what makes such efforts sustainable.  It’s my hope that, through my work and that of my colleagues, we’ll build a science around mass collaboration that will not only help ensure that Wikipedia remains a valuable resource, but will also make new, successful Wikipedia-like communities easier to create and maintain.

Are than any groups or organizations that you’re involved in?

As an academic, my closest org is the loose knit, international group of researchers who are interested in studying the same phenomena that I am. We meet a few times a year at conferences to discuss our progress and publish our results. These other researchers have become like family to me.

Being a Wikipedian consumes most of my energy outside of research. I spend most of this time writing software tools for Wikipedia editors (e.g. Wikignome is a tool that lets you edit a sentence while you’re reading it) and a small amount of time actually making edits myself — usually in articles about computer science, statistics or biology.


U of M’s new GroupLens prof wins 2013 Yahoo ACE Award
U of M’s Grouplens Research Lab Goes Cyclopath
U of M spinout Ninja Metrics raises $2.8m