Q&A: Townsend talks research, goals for semester

Editor's note: This interview has been edited for length and clarity.

Richard Townsend, new to the Department of Computer Science since this fall, wants to make COMP11 (Introduction to Computer Science) and COMP40 (Machine Structure and Assembly-Language Programming) more effective by understanding why the courses are built the way they are and adding more transparency. Townsend sat down with the Daily to discuss his research, goals for the semester and how he first got involved with his barbershop quartet, Madhattan. 

Tufts Daily (TD): What brought you to Tufts?

Richard Townsend (RT): So I was getting my Ph.D. in computer science at Columbia … in 2013 … By the end of my undergrad, I was like, "What the heck am I doing with my life?" I knew I wanted computer science in my world, but I didn't know how. At that point, I saw two main paths – either to go work in the industry and work for a company, or go into academia and keep learning, you know, get a Ph.D. and then work at a university or college. At that point, all I really knew was that I liked talking about computer science and learning about it more than just programming for hours. So based on that, it seemed like staying in academia was a good fit. Over time, I learned that teaching was my favorite thing about it so I started looking for teaching-focused faculty positions. When I was gearing up to defend my dissertation last year, Tufts had an opening. I went to Oberlin [College] for undergrad which is like super liberal and artsy and it seemed, just based on the materials that I saw that Tufts — at least in the Computer Science department — had a similar vibe to liberal arts, especially with its accessibility to students and its focus on learning for learning’s sake as opposed to what job is it going to get you.

TD: What made you decide that you wanted to do computer science?

RT: I took a class called Intro to Java. It was a single trimester course in my junior year of high school and I just found that programming aligned really well with how my thought process works. At that point, I thought that computer science was just learning how to program really well. And I was into that. So I decided I would major [in it] when I got to undergrad. But then every class I took continued to divorce me from that idea, and I realized that programming is a tool. One of my favorite analogies says that "a computer to a computer scientist is like a telescope to an astronomer — it's a tool that we use to perform our craft as opposed to being the craft itself." That blew my mind and the deeper I went, the more I got hooked on just how everything works together.

TD: Could you describe your research a little bit?

RT: My subfield of computer science is programming languages and compilers. So when we write a computer program, the computer can't actually understand it. We either have to use another program to interpret it or use another program to translate it into a form that can actually be executed by the hardware. That form is usually a bunch of zeros and ones that look like gobbledygook to us. The hardware that we have in our machines is pretty good at doing anything, but if you designed a piece of hardware to do one specific task, it would probably do that task much better than the general purpose process of hardware that we have in our machines. There's this burgeoning field called special purpose hardware, where instead of doing everything pretty well, it does one thing unbelievably efficiently. This is super important in like our phones because if we actually tried to utilize all of the computing power at its maximum potential, our phones would melt in our pockets. So instead, hardware designers have looked at trying to build more special purpose hardware and simplify the process of designing and testing those things. My research went into how … [we can] simplify that process because traditionally, hardware designers have to think like hardware — think at the level of zeros and ones and wires and electricity. But if you have a bunch of numbers and just want to sort them — which is a typical problem — and you have to figure out how to manipulate bits and wires to complete that task, it is much harder than writing another computer program that just does it for you. So my research goes into figuring out whether we can take a programming language that humans are more interested in working with than just zeros and ones and wires and, given a program, somehow translate it into blueprints for a hardware circuit that does the same thing. In other words, a special purpose piece of hardware that does the same thing as a software program.

TD: Are you going to continue researching while you're at Tufts?

RT: Almost certainly not: I'm a teaching professor. There are no research expectations or requirements for me. I was looking forward at the end of my Ph.D. career, [to] just leaving the research behind and getting to focus on how I teach and the pedagogy.

TD: What are you teaching now?

RT: Right now, I'm co-teaching COMP11 (Introduction to Computer Science) with Megan Monroe. I am also teaching COMP40, which is [about] the machine structure and assembly level programming. That class has two goals. One is to take our CS (Computer Science) folks, or anyone who's ready to go down this route, from being newer programmers to learning how to program in a much more methodical manner. The other is getting students to understand how the machine actually works. So, diving more into hardware and how memory is actually built.

TD: What excites you most about Tufts?

RT: The [computer science] department is growing at an insane rate. Definitely in terms of the demand, and because we have such growing demand, we're also hiring a lot of people that have more specialties. So one thing I'm interested in is having the freedom eventually to teach courses on niche subjects that align with either what I'm most experienced in or a course that I think would just be interesting. So I'm just excited about the potential for more growth in this department, both in terms of courses offered, students taking them and faculty to teach them.

TD: Do you have any goals for the end of the semester?

RT: They're specific to my courses. One of the reasons that COMP40 is so notorious is because the assignments are like these beasts. With one assignment, the directions might be like 14 pages long and 50 or 100 tests are run on student assignments. My goal is to really understand the class framework that's existed for years so that I actually know how we are assessing students in this course and can determine whether it aligns with the learning objectives. I also want to be more transparent with students so they have a sense of how well they're doing. In COMP11 last semester, Megan [Monroe] mostly, and I helped rewrite basically every homework and lab. Some of them were very successful and some of them led to dumpster fires, so I'm really excited about shoring up the weaker points that we found over the course of last semester and hopefully just continuing to improve.

TD: I read that you were in a barbershop quartet.

RT: I am.

TD: Are you still doing that?

RT: I am. I go back to New York about twice a month. One of those weekends is for an intensive weekend of rehearsing with my quartet.

TD: How did you get into doing that?

RT: Since like eighth grade I was obsessed with collegiate a cappella music. Initially, I wanted to go to college so I could sing a cappella music. So when I got to Oberlin, I joined the Obertones. A member of the Obertones was super into barbershop music and one day after rehearsal he was like, “Hey, you want to sing a tag?” The tag is the end of a barbershop song. We did some tag singing and then he asked if I wanted to be in a quartet with him and some other members of the a cappella group and I was like, “Alright,” but I was still much more into the college a cappella scene than the barbershop scene. When I moved to New York, I went a year without singing and I missed it a lot. Then the same person who first got me in a barbershop told me about this great barbershop chorus in New York called the Voices of Gotham. Eventually, I went to rehearsals and they were unbelievable. I joined them and was a member … for four years before I moved here. As part of that, I also formed a quartet with members of the chorus who were also members of the Obertones at Oberlin. We’re called Madhattan, and I’ve been singing ever since.