A recent opinion piece by Kirk Goldsberry, visiting scholar at Harvard University, gained a good deal of attention because it advocated for a return of a geography department to that august institution. Sensors & Systems (S&S) special correspondent Matteo Luccio spoke with Goldsberry about his background, about the need to reveal spatial patterns, and about the importance of geography.
S&S: Did you like to look at maps when you were a child? What’s the first map you remember?
Goldsberry: The first map I remember looking at was the Rand McNally road atlas in the back seat of the family car. I think we were driving through Belzoni, Mississippi. I remember studying its pages to figure out where we were on those lost highways and where we were going and what we were going to pass through next and trying to understand how these maps would unlock the secrets of the world. I think that many Americans primarily associate maps with highways and road trips, and I guess I started that way too. Nowadays, kids just play Candy Crush Saga in the back seat and people don’t even go to Belzoni.
S&S: Why did you decide to study geography?
Goldsberry: I was an undergraduate meteorology major at Penn State and was required to take some geography courses. I took a landforms class with Cindy Brewer and that turned me on to geography and I never looked back. Cindy had these hauntingly gorgeous slides of landscapes and she had such a conversational style in the lecture hall—it blew me away, and she became my favorite professor at Penn State. I signed up for all of her classes, and she teaches the best cartography classes in America. Years later, as a professor of cartography myself at Michigan State University, I was trying to copy her style, but was never quite as good. So, the decision was accidental in the sense that I stumbled into a geography course as a requirement for my original major at my university.
S&S: What are some of the elements of the “invisible terrains” that “shape urban living” that you write about? How does one map them and make them visible?
Goldsberry: I’ve always been fascinated by geography’s ability—and particularly contemporary cartography’s ability—to unveil invisible spatial structures in our cities and in the world. There are these hidden terrains that we all traverse on a daily basis. The ones that interest me most in that sense, are patterns of accessibility to different services, such as health care services: depending on what part of the city or of the state you live in, you have to drive much further to access basic health care. It is similar with access to nutritious food, to produce. In all American cities, there are heterogeneous spatial structures that affect us all dramatically day to day, but are really invisible in the natural world.
One of the big things that mapping can help us do is to reveal those spatial patterns, those invisible terrains, that shape urban living. We are talking about 1) the configuration of opportunities on the ground—whether it is grocery stores, churches, or Burger Kings—generally represented as a series of points and 2) about a model, which could be as simple as including 1-mile Euclidean buffers or as complex as time-cost or even more complex than that. We can estimate which homes are within 5 minutes or 20 minutes of the Methodist Churches in Belzoni. Since geographic accessibility is a ubiquitous, gradually changing phenomenon, it makes sense to kind of conceptualize that as terrain and depict it using isolines.
S&S: What are some of the most interesting research questions in geographic visualization today?
Goldsberry: About ten years ago Mark Harrower, who was one of my TAs at Penn State and became a really good professor of cartography himself, said that the bottleneck in terms of visualization was no longer computation but the cognitive abilities of the observers or the users. That’s even truer today and we haven’t really come as far as you’d think in terms of streamlining these things or optimizing communication from a cognitive perspective. For the first thirty or forty years of the field, in terms of computational cartography or visualization, the challenge for Tobler, Clarke, and Dangermond was really just making these maps, building these analytical devices, making data-driven things with the computer.
That was hard, but they made incredible strides, and we’ve all benefited from their toil. That’s not hard anymore. Now the challenge is how can we optimize these devices as communicative forms. Unfortunately, Harrower moved on to other pursuits, and I don’t think many researchers today are making the kind of strides we need in that domain.
S&S: How can the applications of geographic technologies help us better understand the world around us?
Goldsberry: I don’t think there’s really anything new there. Maps continue to play a huge role in helping “us” understand the world. Fundamentally, nothing has changed. We have a few new buttons to push, but I don’t think we should be seduced by those technological changes and pretend that these recent shifts are more significant than they actually are.
What academics are now labeling “GIScience” can help by getting spatial thinking into the minds of more human beings. If there’s one valid criticism of mapping that I think is very fair, it’s about the “us” that you mention. We are a small group and availability of mapping technology has never been widespread—GIScience has certainly not really changed that unfortunate truth. In most countries, most citizens have never even seen things like Google Maps or other technologies so easily accessible to “us.” So, it’s a great question with many answers. The one I’m going to give you has to do with spreading the word, spreading access to the technology, getting it into more hands, helping more human beings achieve basic spatial understandings.
S&S: How does geographic illiteracy differ from innumeracy or other gaps in fundamental education?
Goldsberry: I think there’s a real problem when people don’t know where stuff is, and, more broadly, when they can’t reason spatially or don’t think spatially in their day-to-day lives when confronting huge problems like climate change and public health and warfare. All these things are blatantly spatial and take sound spatial reasoning to understand, yet we’re not teaching students at any level, at least in the United States, explicit spatial-reasoning skills and we’re certainly not teaching most people how to conduct contemporary spatial analyses. So it differs in the sense that it’s kind of ignored relative to those other forms of illiteracy, but it brings its own unique brand of consequences.
It’s really strange to not see more geography classes common in American universities, where our “best and brightest” students are going to learn so that they can go help the world. I don’t really like to use the word “illiteracy” to describe that phenomenon, but prefer to say that it is underrepresented in the course catalogs at our best universities. I do think that we’re seeing consequences of that in government, in industry, and in the media.
The world of data visualization provides an interesting example; I don’t like that term “data visualization.” I think a lot of that activity has to do with representing worldly phenomena and things in the world in a way that triggers spatial reasoning in our minds, not “visualizing data” or things that are zeros and ones on computers. Now, that might be nit-picking, but it’s a symptom of people improperly elevating the role of computation over the underlying goal of understanding something new about the world. Geographers like Minard or Bertin have been thinking about these issues for a very long time and have made really fundamental contributions that are not exposed to the masses. So, that’s one of the things that I think is different. We all know what happens when you have never heard of the wheel.
S&S: What motivates college students most to study geography?
Goldsberry: I don’t know what motivates most college students to study geography. I’ve seen many types of students pass through geography courses. It is depressing that most people like myself find geography by accident. I think that’s partly because it’s not thought of as a major major, if you will, in American universities, and that might be in part because it’s not in the high school curriculum anymore. Once college students understand what geography actually is as a discipline and what it can be applied to as a reasoning process, they are at least motivated to get to know geography a little more, whether they want to major in it or not. Once students realize that geography can help you understand everything from climate change, to electoral distribution, to redistricting, to urban planning and public health, that’s when we see the motivation. That’s when you get that “Aha!” moment. Geography professors and administrators at universities need to do a better job of getting the word out there, and that’s not easy. I’m not saying that I could do it better, but I do think that’s what the challenge is.
S&S: What is your best argument as to why Harvard, or any university, should have a geography department?
Goldsberry: Every university should have a geography department because thinking spatially, reasoning spatially, quantifying things spatially, and communicating things spatially are ridiculously vital in the “big data” era. There are many disciplines that can help you learn to do some of that, but I believe in my heart that geography has a huge seat at that table. I haven’t encountered any other departments at any university where our kind of emphasis on spatial reasoning is nearly as strong, and so as we race headlong into the big data era and all these things have huge spatial components, I think there’s an easy argument to make about why we need these students to have geographic education. Put another way, what is your best argument as to why these departments should not exist?
S&S: Should a college geography curriculum now differ fundamentally from what it was 30 or 50 years ago? If so, how? Why?
Goldsberry: Yes and no. I think there’s a tendency in universities to treat GIS as if it is a technology and I think that’s where we run into a wall and why administrators start to doubt us. Universities are not trade schools. The key is to link the emerging technologies with age-old reasoning techniques—the kinds of formulations that the geographers have been doing for a very long time. So linking Ptolemy to John Snow’s map to GIS, for instance. The key is saying that these are new technologies but that they are only helping us understand age-old questions. Where I see a lot of the curriculum running into difficulty is when it becomes about pushing buttons on a software platform and not about problem-based learning. So, it should obviously differ in the sense that there’s more technology in the classroom, but it should not differ so much that it’s not the same basic principles that guided geography classes fifty years ago or a hundred years ago.
S&S: Is cartography half-way between graphic design and geographic science?
Goldsberry: Cartography is informed by graphic design and by geographic science, but I don’t want to put some sort of linear path between those two paradigms and say that it is halfway along it. The best cartographers are very potent communicators in part because they know how to handle data, they know how to reason spatially, and, most importantly, they know how to communicate that graphically in a way that is very congruent with human intelligence. Maps have always been some of the world’s more special documents. They say that a picture is worth a thousand words; well, a map is worth millions of words, in some cases. Trying to locate cartography in the context you have in this question is troubling to me. I do think that a good cartographer should have some graphic design education. I do think that a good cartographer needs to have statistical, scientific, and computational training as well. In that sense, cartography can be informed by graphic design and science. However, to say that it is located neatly between these two is a little simple. But it’s provocative, I’ll give you that.
S&S: How have GIS and other digital tools changed the art and science of cartography?
Goldsberry: When I was a young cartography student, in the late 1990s, my professor, Cindy Brewer, was fond of saying that we were only two or three years removed from that same cartography class, Geography 321, being a very manual process, almost entirely a manual drafting-like procedure. Then, in the late 90s, it entirely switched to computation and the tools of the trade literally had nothing in common. The stratagems of the trade have not changed—that’s a point I’d like to emphasize—but the tools of the trade have changed dramatically. In that sense, I believe that GIS and other tools have changed the art of cartography in terms of its application. But the goals and the reasoning behind it are timeless. There’s a lot we can learn from those cartographers—whether you’re talking about Raisz or Bertin or Judy Olson or Waldo Tobler—who all gave us these fundamental contributions that remain relevant regardless of digitization. The core rules have not changed because we have computation.
S&S: In your CV, you wrote that you “[h]elped initiate and lead the redesign of the cartography now featured in Delorme’s most popular software releases, Street Atlas USA and TopoUSA.” What was behind that re-design? What were you trying to achieve?
Goldsberry: Back in 2000, DeLorme was trying to put out map atlases on CD-ROM and it was a pretty fascinating time to be a cartographer. What we’re looking at in terms of Google Maps in 2013 is the best cartographic masterpiece of the last fifty years. It’s a wonderful product. But it didn’t just show up overnight. There was a time about 15 years ago where a lot of companies were racing to build digital atlases—DeLorme was one of them. It wasn’t clear who was going to win back then, but here in 2013, it’s pretty obvious who won that race.