Skip to Content

  • A
  • A
  • A


News Room

Poetry of Data: An Interview with Miriah Meyer

September 02 2016 - 2:16 pm

This is the fourth installment of a series of conversations with Utah women building disruptive technologies. I’m Kimberly Zhang, and this summer, I set out to speak to the women behind the business, research and ideas that are changing the world. They share their work, thoughts and advice. Note the opinions expressed by interviewees do not necessarily represent those of GOED, but they do promise to be interesting.

I sat down with Dr. Miriah Meyer from the University of Utah School of Computing in the Warnock Engineering Building on campus. We talk data visualization and effective group dynamics.

What brought you to Utah, and why did you stay?

I did my graduate work at the University of Utah’s School of Computing. Most schools don’t even have a single person that does data visualization, but at the U we have six. That’s what brought me here originally.

I left to do my postdoctoral work at Harvard. When I was looking around in the job market, I came back, because I wanted to work in a larger group again. The department here has opened up so many new opportunities to grow and rethink what computing means. Is computing just operating systems and networking? No, it’s much bigger than that. We have a project in digital humanities that combines fine arts students and CS students. There is a faculty member in this department who does joint teaching with a sculpting professor to create art that moves. It’s exciting to be part of a department that is growing, changing and trying to address this question of “what is computing?”

Miriah Meyer. Photo: Ryan Lash

Miriah Meyer. Photo: Ryan Lash

Tell me about your research in data visualization.

My group does data visualization in many different application areas. In my post doctorate, I did a lot of work with biologists. Biology is a really exciting field, because it’s become very quantitative and there’s a lot of data. As a visualization person, there are tons of opportunities to create tools that have real impact in the scientific process.

Since coming back to Utah, I’ve really expanded into many application areas. I have a project with poets on visualizing poetry. I have a project with a geographer who is working with UTA (Utah Transit Authority) on understanding the intersection of socioeconomic issues with transportation. This broad spectrum of projects is why I love visualization and computer science. I’m someone who has a tendency to get bored easily. In what I do, I always get to work in some new area. No matter the field, there is always going to be a computing problem.

The style of work we do is unique, because of our focus on data visualization as a design process. Our work is very deeply collaborative. We spend a lot of time with people trying to understand their thinking processes, culture and intuition. That’s really important to designing effective tools and incorporating user centered design. We incorporate rapid prototyping from sketching on paper to quick software development. It is really holistic. I’m an engineer by training. I still like rules, disciplines and guidance, and we’re trying to apply that concept to very messy, human-centered processes.

Is there a step-by-step process for figuring out what tools are needed to interpret the context of the data?

That’s the cutting-edge research in our field. My group has published a couple of papers to put some structure and guidance around this messy design process. That’s the main thrust of our work. What we do is what I call “data counseling.”

I feel like my role as a visualization designer is to start asking questions. It is like this counseling process where someone comes in and says, “I have a problem,” and you have to start unraveling neurology, systems biology or poetry. I start asking questions like “why do you need to visualize?” that to get to the deep-seated goals.

Then, there’s this process called operationalization, which is about taking those real world goals and questions and making them concrete, explicit and linked to data.

This process of data counseling often uncovers things someone wants to do that they often times aren’t even aware of. We take that and try to translate it into a set of tasks and links to data we can actually design around.

How does visualization help retain the richness of massive amounts of data?

The argument I usually make is that visualization is particularly good for data richness in ways that some other approaches aren’t. I use the U.S. News and World Report college rankings as an example. It’s just a ranked order list of schools. Stanford is better than Harvard, right? It turns out that they collect lots and lots of data about schools, and then that data goes into some numerical model that weights everything in some complex way and spits out a ranking. Visualization allows you to see more than just a list of stats. Through interaction, you can use visualizations to tailor the data to answer your own unique questions and form your own decisions.

You started in astronomy and astrophysics for your undergraduate studies. How did you get to this point now?

I’ve always loved math and science. Astronomy was a cool way to study those things, but at some point I decided that I didn’t want to study one wavelength for the rest of my career. I didn’t know what I was going to do, so I traveled for a while, eventually ran out of money and had to get a job. I got a job in programming. I didn’t know what in the world I was doing, so I started taking some basic CS classes and fell in love. I took a computer graphics class and one of first things the professor showed were some beautiful 3D interactive visualizations of the Orion Nebula cluster, which is the cluster of stars I did some of my undergraduate research on. I was like, “Oh man, computer science is the way to be involved in science while still making and building something.” That’s the part I really love.

Your toolbox seems pretty interdisciplinary. For someone entering the field today, what sort of skills would you recommend learning or exploring?

A lot of people can create visualizations without any sort of computational or technical background—designers, data scientists and journalists. From my experience, you have to be familiar with data science. And if you really want to create novel, interesting and interactive visualizations, having a some programming experience is really critical.

How do you see data visualization changing in the next 30 years?

Like you pointed out, visualization is popping up on all the news media sites. People talk about the importance of visual literacy. I have colleagues who are bringing visualizations to kindergarten and elementary schools, in order to understand the cognitive limits in how people learn about visualizations. It’s going to be part of our communications lexicon. I personally hope that computer science becomes a part of general education—that more and more people will have the basic skill set to think computationally, to think algorithmically.

Data indicates that the number of women in CS is decreasing. Why do you think that is?

That’s a conversation that CS departments are having all over, but it’s really complicated. Based on my own experience in the field, it’s two things. For one, there is this perception of the CS culture. The perception used to be a trench coating-wearing unhygienic gamer sitting in a cubicle, and it’s not just women who aren’t interested in that. There’s also this other new cultural stereotype, and that’s the “brogrammer” being perpetuated in Silicon Valley with all the startups—it’s all these 27-year-old bros who are about being at work all night. There’s a lot of people who don’t have interest in that.

Another part of it is that in the academic pipeline you see the number of women decrease at each level. One of the reasons is confidence. To be honest, I’m someone who has always struggled with imposter syndrome. I’m like, “One of these days they are going to find out that I don’t know what I’m talking about.”

Some people might argue that gender diversity in STEM isn’t important. What would you say to that?

There are studies on how teams function better with women. I’ve had colleagues who have said to me, “When you come into a meeting, the conversation is different. It’s just more respectful and open and that’s really important.”

We need to learn how to work in teams, and women change the dynamic. There’s this great piece in the New York Times on workplace culture. Google looked at teams that succeed and those that didn’t. They brought in social scientists to look at metrics like time, effort, et cetera, and what they came up with was the idea of psychological vulnerability. The successful teams had high psychological vulnerability. They share things about themselves. They had scenarios where the managers told something very open and honest about themselves and it supported better collaboration, more sharing of ideas and listening, and more teamwork.

It put some words to how I work, which is this very gendered female thing. I am very honest with my students about where I feel like I’m going through a midlife and mid-career crisis. I joke with my students about that. Because of that openness, it changes the dynamic. The students open up and work with each other more.That goes back to creating a situation where people feel empowered and people don’t feel like scared to say what they think. That goes to say that not all women are “warm and fuzzy,” and some men are “warm and fuzzy.”

What solutions are there?

It’s a real problem. It’s a big thing here at the U. We try to have things like “Women in CS,” where the women faculty take undergraduates out to lunch once a semester and things like that. One really important thing in CS done on a national level is the Grace Hopper software conference. It’s a huge conference—11,000 people and almost all women. The first time I went was in graduate school and I remember thinking, “What? These are all computer scientists?” It blew my mind. That’s something we send undergraduates to every year. Just to make it feel less invisible.

I ran into one of my graduate students in the bathroom and she said, “That’s so strange. I never thought I would run into my adviser in the bathroom.” It was even a bit of an experience for me. Also, I have a two year old, and I try to include him. When I introduce my group, I always introduce my son as our lab manager. I try to include what’s going on at home. Just to normalize it and say you can have a career and family, because I certainly didn’t see that when I was young and going through grad school.