By Lotem L. Loeb

Fifteen Questions: Elena Glassman on Human-Computer Interaction, Freestyle Wrestling, and MIT Dorms

The human-computer interaction expert sat down with FM to discuss software design, the importance of bicycling infrastructure, and her time competing in women’s freestyle wrestling.
By Sage S. Lattman

Elena L. Glassman is an assistant professor of Computer Science. Her research centers on human-computer interaction.

This interview has been edited for length and clarity.

FM: You study human-computer interaction. Can you explain human-computer interaction in a “Human-Computer Interaction for Dummies” way?

ELG: Given the pervasiveness of computers no matter where you are, you’re often able to describe a human interacting in some way with a computer. And so anytime there’s that interface, that interaction, then HCI is potentially relevant to studying how the interaction is going, the ways in which we can construct it to go better.

FM: How does human-computer interaction incorporate elements from other fields besides computer science, such as psychology and neuroscience?

ELG: Because there’s many different areas within HCI, people are contributing to the field from places like anthropology, sociology, in addition to computer scientists. But for me personally, sometimes I feel a little bit like I’m asking psychological questions that have been made more specific to the context of a human or humans interacting with a computer or an AI.

FM: Unlike some other CS programs, Harvard’s doesn’t focus much on software industry skills. However, many, if not most, CS graduates will go on to work in the industry. How do you — or do you — think the curriculum prepares Harvard CS graduates to build good software?

ELG: I think what we, as educators in the CS department, care about is basically future proofing people’s knowledge. If I teach you a software engineering course in which you nail Java, that will serve you well as long as Java’s around. And hopefully you got those higher-level concepts that are going to be probably preserved in other languages that take over.

FM: Could you tell me a bit more about “CS 178: Engineering Usable Interactive Systems,” and what you hope undergrads get out of it?

Glassman pulls out “The Essence of Software: Why Concepts Matter for Great Design” by Daniel Jackson.

ELG: The best thing I found so far for thinking about usable interactive systems. It’s not to say that this book is perfect. I think Daniel Jackson has thought really deeply, and he’s created a text that is fantastic to serve as the driver of a conversation that the students and I are having as a learning community.

One of the fundamental insights that came out pretty early in the class this year was that usability is not a function of the object. It’s a function of the person and their prior knowledge in relation to the object.

So something that’s really useful for you because you also use Tik Tok and it’s like, ‘oh, it’s just like Tik Tok,’ might be very difficult for someone who’s never used Tik Tok.

We can make really novel, exciting new usable systems by coming up with unique combinations of pre-existing basic building blocks.

FM: What aspects of human-computer interaction do you think are most overlooked by people who design software?

ELG: My hope for all graduates of any CS program would be that they had watched over the shoulder of someone using a piece of software that they wrote. And I think if you do that once, you will be inoculated from thinking that what is obvious to you will be obvious to a given user. And then you will hopefully — having that memory seared into your brain — bring in user feedback early and often during your development process.

FM: Some of your recent work has been about AI-resilient interfaces. Can you explain what this is to someone unfamiliar with the topic?

ELG: The generalized idea is if you are using an interface, and it has features that are AI powered, we need to, as interface designers, enable you to be more resilient to the AI choices that aren’t right.

In order to enable the user to be resilient, we think that the user needs to be able to notice reliably and have the context necessary to judge well these AI choices that are being made.

FM: You were an undergrad and Ph.D. student at MIT. What, if anything, is better about Harvard?

ELG: Harvard students are very ready for the type of rigor that HCI requires, and that type of rigor is not necessarily deriving some optimal interface. We don’t have the math for that. If we did? Cool! But we don’t.

I feel like a humanities and liberal arts education is an excellent preparation for that type of rigor. I think that notion of rigor as being rigorous, was something that was harder to get across for some students at MIT.

FM: What do you miss about MIT?

ELG: I belonged to an undergraduate living community called East Campus that was kind of for eccentric misfits.

Because we lived in a dilapidated old building that had been around forever, we had decades and decades of student murals that plastered every wall. Every flat surface had been painted by someone, and it was basically a building of art.

Unfortunately the building has been finally slated for renovation, and all the students have been moved out. They are going to basically gut and reconstruct the inside. I’m sure it’ll be more accessible and all these kinds of things. But also, I’m sure that when the students get back into the building, they will not be able to paint, and it will never be the same.

FM: You’ve been tweeting a lot about pedestrians and cyclists. What are your thoughts on the Cambridge City Council bike lane debates?

ELG: Oh, God. Yeah, so, kind of a funny story. During the pandemic, I got radicalized by a really, really wonderful channel called City Beautiful by a professor at UC Irvine. He gave me language to understand the urban environment and what makes it more or less habitable.

I think that, you know, if you look at Google Maps, and you’re like, what’s the fastest way to get from A to B? In most weather, it can be bike.

The more people can feel comfortable enough to bike, the fewer vehicles are on the road, and the less pollution. And the driving experience for those who need to use the vehicle will be better. I think everybody wins with building bicycling infrastructure that is up to the international standards.

FM: I saw from X that you run. Do you use any running specific apps or gadgets?

ELG: I’m on Strava. I have a Garmin watch. The running community at Harvard has been really lovely.

FM: I saw through a Medium post that you used to do competitive freestyle wrestling and judo. How did you get into those sports and what drew you to them?

ELG: Every summer during undergrad, I rented a room in the Zeta Beta Tau fraternity in Brookline.

There were only a couple magazine subscriptions that the house had. One of them was ESPN, and they had a spread in the lead up to the 2008 Olympics during my senior year of Team USA and someone representing each sport. There was this woman who was wearing what looked like a leotard — like gymnastics, but it was subtly different — and it said wrestling underneath it. And I had the very naive thought of: I could take her. I’d never wrestled before. She didn’t look like the Hulk. She looked like the girl next door.

So I invited my roommate, who I’d just met, to be my training partner and for us to walk on, make use of the fact that I had a fifth year of NCAA eligibility. So we together joined what was then the men’s Division III MIT wrestling team.

We occasionally got matches at men’s collegiate tournaments, which were pretty short matches for me, as you can imagine, because these were men who were wrestling for many, many years. But I was really into just learning how to do it better. As soon as that season was over, I would drive hours in each direction to train with other women. I would drive and compete in tournaments in Canada.

I never was objectively particularly good. But I was very proud to kind of be a faculty sponsor of the Harvard Women’s Wrestling Club, which is now doing fabulously.

FM: What sport did you play when you were in college?

ELG: Freshman year, I was on the winter track team at MIT, and it was a terrible experience. That started my five-year NCAA clock, and my final fifth year was wrestling, which was a much better fit in terms of interest.

FM: Recently, there’s been a flourishing of startups hoping to reinvent the way we interact with technology, such as the Humane AI pin or Harvard dropout Avi Schiffman’s wearable assistant Tab. What do you think of these new developments? Do you think they’re preferable to traditional devices with screens?

ELG: One of the unavoidable challenges of voice as a medium as opposed to a screen is that it’s very difficult to communicate to the user of a voice assistant what the voice assistant can and cannot do. There’s no information sent. There’s no different buttons you’re gonna look at and press where those buttons have descriptive names.

I think those sound like cool startups, I’m sure. Hopefully, they’re doing a lot of user testing.

There’s this apocryphal story of when people were developing the first computer. They went to the typists — probably unfortunately all female — who were just typing up stuff. They were like, “What would you like?” And they’re like, “Can you make me a better typewriter?” There’s this notion that people can’t imagine what would be possible for them to ask for.

Part of the job of a designer is to both identify needs that people are aware of that you can help with, as well as things that you know because you’re so familiar with the design material. It’s possible to lower an obstacle that people currently face to achieving a goal that they just didn’t even consider. Like, I don’t wake up and think, “Maybe I’ll try flying today.” It’s just not part of my world of possibilities. But if someone is like, “Gosh, I actually do know how I could get enough lift by attaching something to you. I can lower the obstacle to you flying.” Well, suddenly, flying becomes an alternative commute option instead of biking!

FM: Tell me about a beautifully designed piece of software you use that people might be interested in.

ELG: I use Superhuman for my email often. Not exclusively, but I think it’s pretty beautiful, and I think it’s pretty functional. It has a lot of hot-keys. You don’t have to move your hands off the keyboard, and I really appreciate that.


— Associate Magazine Editor Sage S. Lattman can be reached at sage.lattman@thecrimson.com. Follow her on Twitter @sagelattman.

Tags
Conversations