Brown CS News

“We don’t have a robust theory of humans,” Will Crichton says. But he’s working on it. 

Formerly a Brown CS postdoctoral researcher advised by faculty member Shriram Krishnamurthi, he returns to the Department this fall as assistant professor. He’s one of two recent hires in the multi-year CS With Impact campaign, our largest expansion to date.

Will defines his research goal as empowering people to use the full potential of computers through programming: “How do you build complex programming tools in a human-centric way?  How do you take a body of knowledge about how humans learn from cognitive science and other disciplines and use it to make informed decisions? One of the things I love about my discipline is that it’s full of tech-savvy people who are all interested in the same sort of story.”

As the child of journalists who were early adopters of teletype machines and text editors, Will’s first exposure to computing was through Microsoft’s Encarta multimedia encyclopedia and the Zoombinis educational puzzle series. At sixteen, trying to avoid a job in a supermarket, he found a web design firm willing to hire a teenager, and began working with HTML, CSS, and JavaScript.

During his four years at Carnegie Mellon University, Will left behind what he calls his “high schooler’s myopic view of CS” as being just about web development and startups. He explored many areas of computer science but became increasingly disillusioned with the prospect of a software engineering job: “I wanted autonomy, being able to work on problems I cared about and trying to change the world, not being beholden to a bottom line. Software as a public good. It was all too easy to create systems that I knew would disappear into the ether, and I was getting more and more interested in the community I was working with, the human aspect.” 

Halfway through a doctoral program at Stanford University, Will decided to make a major change, moving his research interests much closer to programming languages and CS education. Will uses the term cognitive engineering (his research group at Brown is the Cognitive Engineering Lab, and he’s currently recruiting PhD students) to reflect his interdisciplinary approach, applying theories of human cognitive capabilities and limitations to understand how computers can best provide cognitive support, building practical systems used by hundreds of thousands of people worldwide.  

Why engineering? Will’s explanation is that the essence of an engineer is someone who constructs things based on reliable knowledge of what they’re building: “They understand material properties, they have good theories about them and can be very, very confident that their bridges won’t collapse. The word ‘developer’ is used more for programmers these days, but I think we can be engineers if we have the right training and mindset. I don’t feel like an impostor claiming that aegis. Programming theory is an important part of software engineering: I can show you a mathematical proof that my software will do certain things, and I want to bring that style of engineering to more human-facing problems.”

By nature, it’s multidisciplinary work, especially in the area of CS education, where Will sees drawing on cognitive science as absolutely necessary: “For a long time, people had a tacit theory of learning programming languages that didn’t ignore the human aspects entirely but treated them intuitively: ‘Programming is hard because I have to type a lot of characters, so I want to minimize that.’ Programming is this enormously complicated thing, and yet we’ve been writing code based on intuition since the dawn of CS. Cognitive science gives us a more precise foundation: if we ask people to debug this particular bug, we can expect it to be hard in these ways and we can bring in these tools to work on it.”

Improving something that’s more than thirty years old speaks to Will’s ambition, and a bold statement about the popular PDF document format (“it’s the end of portable documents”) on his social media account prompts a question: does he see himself as an iconoclast? 

“I don’t think so,” he laughs. “As an undergrad, I was predisposed to grandiose claims about the nature of how things should be, but now I think I take a pretty measured view. When I wrote that line, I thought the PDF format was unredeemable, and I still believe that.”

“Fundamentally,” says Will, “you need a bit of zealotry to change the world: ‘PDF needs to die and I have a solution’ instead of ‘the document experience could be improved’. Of course, you need meaningful discussion and a nuanced solution, not zealotry alone. But if you want funding, for example, you need people to believe that your research is valuable. I’m in a space where a scattered group of people around the world care about what I care about, but there’s no hype wave. If I’m trying to effect a change on the future, part of my job is to get people excited.”  

What’s one of his favorite techniques for creating excitement?

“I appreciate a bit of levity,” says Will. “Play is an important part of life in academia, and I can’t ignore the importance of aesthetics. When I design a program, there are constraints, and it needs to be accessible, but if I can add fun without losing functionality, more people will enjoy it. I care about that, because it helps them engage more deeply and philosophically. Some software is very utilitarian, but if it’s human-facing, I want to see an element of play.”  

“Looking more deeply,” he says, “when I think about CS education, it’s easy to take a non-emotional approach. Emotions are difficult, but it’s imperative to design for them. An introductory CS course has two jobs: to teach you about computing, but also to spark your passion. If someone comes out of an intro course and I see that they have a spark, that’s more important than whether they know what a for loop does. That motivation can last a lifetime.”

Fun for Will personally means lots of reading (Vernor Vinge, Greg Egan, Peter Watts, Ted Chiang), tennis, hiking and being in nature, and recently, maker spaces where he can take part in woodworking, metalworking, and computational fabrication.

Looking ahead to his new role at Brown CS, what excites Will most? Given his earlier mention of hype waves and zeal, the answer is unsurprisingly grand. 

“One of my favorite things about Brown CS,” he says, “is our strong presence in graphics due to pioneers like [Thomas J. Watson, Jr. University Professor of Technology and Education and Professor of Computer Science] Andy van Dam. I want to create a similar excitement around human-centered programming. I want us back at the forefront of a wave. As one example, I’m interested in an alternative to C++ called Rust that’s representative of a new era of PL technology. Very few universities teach it, but I use it for a lot of my research, and I want to help people take advantage of it. I want to tell them that they’ll be far better served with Rust for many reasons: not only is it easier to pick up, they’ll be more productive in the long term when writing software in the real world.” 

“A bunch of us here,” Will says, “are interested in human-centered programming as a real feature of Brown CS. I don’t mean as something that we need to push across the entire department, but across academia, there isn’t really one best place where students can go for the human factors of programming – a world center. I’m interested in that. What I’m most excited about is our collective approach. What if we could make Brown a world center for the human aspects of learning programming languages? There’s no better place than Brown to do it, and there’s a big, open space for us to make a name for ourselves.” 

For more information, click the link that follows to contact Brown CS Communications Manager Jesse C. Polhemus.