Interview with Dr. Marilyn Tremaine, Part One

Dr. Marilyn Tremaine, former Research Professor at Rutgers University and recipient of the Association for Computing Machinery (ACM) Special Interest Group on Computer-Human Interaction’s (SIGCHI) CHI Lifetime Service Award in 2005, is an important figure in the founding of ACM SIGCHI. She helped organize one of the first Human-Computer Interaction (HCI) conferences in America in 1981 in Ann Arbor, Michigan, where she was a professor at the time. Later, she co-chaired the 1986 CHI conference, served in various positions on the SIGCHI board over the years, including vice-president of communications and president. In addition to her work with SIGCHI, Tremaine has done extensive research in human-computer interaction (HCI) on topics such as incorporation of HCI into software development processes, collaborative software, stroke rehabilitation, global software teams and educational games that improve STEM discipline skills.

In the following interview, she discusses how she entered the nascent human-computer interaction field and the founding of ACM SIGCHI.

Entering the HCI Field

How did you learn about the human-computer interaction field?
In 1974, I wanted to go to graduate school and was looking for HCI. It didn’t have a name, but I wanted to study how to make interfaces better for people.

I was a project manager—I had moved up in the computing world—working at the Lawrence Berkeley National Laboratory. I was running very large projects and developing software for the US government.

I had two problems with my career. The first one was that I was female, and it’s hard to say what it was like in 1974 being female in a science field. There weren’t other women—there were women at the National Lab, but they were secretaries and staff members. They were not scientists. Second, I was also at a level where I was the only one without a Ph.D., so I decided to get a Ph.D.

I built interfaces for people in government to use. My biggest project was building the system that managed the U.S. Navy’s recruit education system. I wrote the database system and then wrote all the interface software so people could manage their courses. We constantly had people entering data, and I would go up to the Naval center and talk to some of the people in Mare Island. They were obviously women, and they were actually quite smart, and they were crying. They were crying because they couldn’t use my software.

I hired somebody with a degree in English to help write manuals. I asked a whole bunch of people to help, and then I decided the only way I was going to solve the problem was by going back to school and studying and figuring out how to solve it myself.

The Annenberg School of Communication had a program in Humans Interacting with Computing—it was called the HI-C program—and I applied there. So I actually went into a program in 1975 that had just started, and which was interested in improving computers for people. So there was a program even back then—pretty amazing stuff, huh?

Xerox PARC

We had a faculty member at Annenberg who was hired from Yale named Jim Carlisle. He was a real gung-ho person, and he brought in speakers for the HI-C program. One of the speakers was from Silicon Valley, because things were starting to happen out in Silicon Valley then. He brought in this young man named Stu Card*. Stu hadn’t finished his degree yet—he was a Ph.D. student at Carnegie-Mellon—and Stu gave a talk about the Cognitive Psychology they were doing at Xerox PARC**.

Now note, I’m transferring all the way over from hardcore Computer Science into Sociology, and it is a big step for me. Then this guy comes in and starts talking about Cognitive Psychology, and I just went, “Ka-bing, that’s it, that’s what I want to do. That’s where we’re going to solve those interface problems.”

I asked him, at the end of his talk, “Do you ever take visitors?” He said yes, and so I called him up on Friday and asked, “Can I come and see you on Monday?”

I drove all the way from Los Angeles to Palo Alto and visited Stu Card and Tom Moran*** for the day. I met everybody in the lab, and they showed me all the things they were doing. I asked all sorts of questions. Then I asked, “Do you ever take summer interns?” and Tom gave me an application form. By that summer, after my first year of graduate school, I was working at Xerox PARC!

What was I working on? I was working on cut and paste, and doing all sorts of studies with people using text editors. They were building Bravo, and Bravo became Microsoft Word, eventually.

I was a weird girl going around Stanford collecting things from people’s wastebaskets. I went to different departments and asked them if they had manuscripts they had retyped, and if I could see the editing marks on the manuscripts. I brought back all this data because we didn’t know how people edited. This research was all new. It was great fun.

The key people in the department were Tom Moran and Stuart Card, and the person in charge of the whole project was Alan Newell from Carnegie-Mellon. He’s an extraordinarily famous computer scientist. Stu said to me, “Why don’t you ask him if you can go to Carnegie-Mellon?”

I did, and he said, “Sure!”

I went back to the University of Southern California and asked if I could go to Carnegie-Mellon for the rest of my graduate degree. They asked what I wanted to do there. I had a few people on my committee that were pretty savvy—one in Computer Science, and one in Psychology, and the one in Psychology said, “Oh my God, you’re getting to go to Carnegie-Mellon? That’s incredible!” and the computer scientist said “Oh wow, you’re working with Alan Newell!”
So the people in the School of Communication let me go, and I went to Carnegie-Mellon. I did my thesis there and stayed there for two years. My thesis was “Why People Get Lost Using Computer Systems”.

* Stuart Card was a Senior Research Fellow at Xerox PARC and is a pioneer in human-computer interaction, recognized with numerous awards in the field.

** Xerox Palo Alto Research Center (now simply known as PARC) is the place where many common software design patterns and pieces of computer hardware were invented.

*** Thomas Moran is a Distinguished Engineer at the IBM Almaden Research Center. He is a pioneer in HCI and coauthored The Psychology of Human-Computer Interaction with Stuart Card and Allen Newell.

The Formation of ACM SIGCHI

If you want to develop the field of Human-Computer Interaction—it is essential to have a professional society. That was very clear.

Interest in HCI was beginning to happen across the country. After I transferred to Carnegie-Mellon, I went over to the UK specifically because there was more going on with the British Computer Society than was happening in the US. I went to places like Sheffield, Cambridge, and Edinburgh, and I gave talks as a graduate student. I’m sure the comment about me when I gave the talk in Edinburgh was, “I’m sure she was talking about something interesting but I couldn’t hear it,” because I had such a bad cold. But I was very focused on making this a field and absolutely sure that a professional society was necessary.

Greg Marks**** was in charge of SIGSOC (Special Interest Group on Social and Behavioral Computing). What that meant is that they were focusing on getting software that handled statistics developed and available in all the universities. This may seem kind of dumb now, but nobody had such tools in the 70s, and if you wanted to do an analysis of variance, you had to write your own program. Then, there was a major company—two eventually—that started producing the software, and so users formed SIGSOC to manage the sharing and specification of the stat packages they wanted.

By 1980 or so, SIGSOC didn’t need to exist anymore. Every academic computing center had such packages and the packages themselves provided substantial support for researchers. Lorraine Borman† was at a small university in Chicago and had become involved with SIGSOC. Because SIGSOC was dying, she worked with ACM, trying to get SIGSOC converted to what would later be named SIGCHI.

In the meantime, Greg introduced her to me, and we decided that we were going to have a conference in Ann Arbor. I think that was either Greg’s or Lorraine’s idea. I was in my first year as a junior faculty member, when you’re not sure if you’re going to get tenure or not. We held a conference on “Easier and More Productive Use of Computing.”

Lorraine came to Ann Arbor to discuss the running of the conference, and I said to her and Greg, “Have you invited so and so? And don’t forget so and so!” They asked, “You know all these people?” I told them yes, and they should come.”

In 1980, we held this conference on “Easier and more Productive Use of Computing,” and either Lorraine Borman or Greg Marks, or both of them, were in charge of it. I got Tom Malone‡ to come, I got Stu Card and Tom Moran to come. Ben Shneiderman# was there, plus a whole bunch of other interested people. It wasn’t a big conference, but its size led to lots of discussion about the problems and needs of HCI. Meanwhile, Lorraine was still working with ACM to get us approved to become SIGCHI.

At the same time, there was a whole bunch of stuff happening in Maryland because of the U.S. government. Ben Shneiderman was a significant part of that. He organized a group called the Software Psychology Society, a group with no president, no secretary, no rules and, no dues. They just met. Anybody who wanted to attend could come.

They decided to hold a conference in 1982, and they advertised it to everybody they knew, just like I had done for the Ann Arbor one. By that time, Lorraine and Greg, together, had gotten the approval to change the society to be one that supported the Human-Computer Interaction discipline.

There were a thousand people that came to the 1982 conference from all over the world. It was amazing! The organizers were blown away. They had expected maybe 200 people, max, to come. We had this meeting about what was to become SIGCHI and the hotel hadn’t taken the bed out of the room, so people were sitting on the bed and sitting on the floor. That was the first SIGCHI meeting, and at that meeting, we voted on what we should call the society. We came up with “SIGCHI” because “SIGCHI” sounded easier to pronounce than “SIGHCI.” The name change was approved later in 1982, and SIGCHI became an official SIG.

**** Information technology researcher at the University of Michigan.

† First president of ACM SIGCHI.

‡ Thomas Malone is a professor at MIT, who started his career as a research scientist at Xerox PARC.

# Professor of computer science at the University of Maryland Human-Computer Interaction Lab and author of Designing the User Interface: Strategies for Effective Human-Computer Interaction.