Telling stories using data: An interview with Jonathan Harris

An interview with Jonathan Harris

Jonathan Harris calls himself a storyteller. But he's also equal parts visual artist, computer scientist, anthropologist, data voyeur, photographer, digital anthropologist, interviewer, and designer. What might be more appropriate is "story revealer" or even “story enabler,” for Harris has the ability to uncover the complexity of human stories by starting with a simple question. His early work started with simple queries about far-flung data sets across the web: OneWord, an interactive presentation of the 86,800 most frequently used English words, ranks words in commonness order into a very long single sentence; 10x10, an exploration of words and photos in the news, stretches our standards about what digital news means; and perhaps his most well-known piece today, We Feel Fine, searches the world's blog posts for occurrences of the phrases "I feel" and "I am feeling," revealing emotion and humanity behind what would otherwise be plain old text. The audience for his designs, then, become the storytellers themselves, using them as a tool to unfold the narratives.

Figure 1: A visualization from the “We Feel Fine” project.

In the wake of the success of We Feel Fine along with Universe, a modern mythology project released at TED in 2007, Harris began to flip his process. Instead of querying the data and asking it to reveal the human story, he's starting at the source—the human kind. Some of Harris' more recent work required him to start with people, offline, in projects like The Whale Hunt, a process that required him to take a photo every five minutes while living with the Inupiat Eskimos in Alaska for nine days (in other words, 3,214 photos). In this and other recent projects, visualization of the data came only after he's sat with humanity for a while.

Access to Harris' particular style of putting together a story is now reaching a wider audience at the Museum of Modern Art in New York City as part of a new exhibition, Design and the Elastic Mind. He was commissioned to do a piece on online dating, I Want You To Want Me, which chronicles the world's long-term relationship with romance by visualizing text in online dating profiles.

Harris has become an unassuming evangelist for a new way of storytelling. He frequently speaks about his work publicly and places as much care and attention on describing his design process as he does with the design itself.

If asked to describe Harris a year ago, I would have said: Jonathan has a way of visualizing the world's data to make it seem more human. But given his recent explorations, if asked to describe him today, I'd say: Jonathan has a way of visualizing humanity, and making it more accessible to the world. And often times, more beautiful.

Over a series of email conversations, I had the chance to talk with Jonathan about what makes a good story, why working alone is better, and where he'll find his next project.

Figure 2: An image/phrase combination from the “We Feel Fine” project.

Given the layered nature of your work, how would you describe what you do?

I just work on things that interest me. From project to project, that can end up involving computer science, graphic design, visual art, anthropology, photography, storytelling, travel, interviews, and probably a few other things too.

Sometimes I try to identify a common thread in my work, and usually come back to “storytelling,” which is a very loose way of encompassing what I do. John Maeda once told me that my work is an anomaly, but that anomaly is a good thing. So I'm sticking to it.

What are the basic elements of a story regardless of the medium? Are there any?

I define “story” quite loosely. To me, a story can be as small as a gesture or as large as a life. But the basic elements of a story can probably be summed up with the well-worn "Who / What / Where / When / Why / How." And feeling. Stories should have feeling, to the extent that they want to be human.

Figure 3: Navigating the “Whale Hunt” project by cast member.

Whale Hunt, one of your more recent projects, required you to move from a pretty human and intimate level with the content to a digital and more detached one in order to create the piece. How did this change in scale and intimacy affect the way you told the story as you developed it?

Well, on the one hand, I could still remember the feeling of the wind on my face, what a big deal it was each time I had to pee (it was -22 degrees out there on the ice), the exhilaration of piloting a snow machine across the moonscape of the frozen Arctic ocean, and the feeling of intense boredom while waiting four days in near silence for a whale to show up.

So I remembered those feelings and wanted to present the content in a way that could best express them. But on the other hand, I suddenly had this distance from the experience, which allowed me to observe its internal patterns and trends, and I wanted to find a way to express those too.

You used constraints to structure that story—the cast, concept, context, and cadence. When you start working with a narrative, do *you* give it constraints so that you can tell a story the way you want to, or do constraints naturally emerge?

Constraints naturally emerge. I look hard at the material, sometimes with the help of programmatic analysis, and see what secrets it holds.

The work that we're talking about is primarily digital. When working with digital storytelling, how do you negotiate between guiding viewers through a narrative and providing options for them to choose their own path?

It's important to do both. The power of defaults will need to accommodate the vast majority of people. But it must be easy and playful to diverge from the defaults and chart one's own course. Here, layering is essential. At any one moment, the range of available options within view should be very small, but it should be effortless (and fun) to get more.

Figure 4: One of the 3,214 images from the “Whale Hunt” project.

Sounds like a significant challenge. How do online narratives usually fail?

They can fail for many reasons. They can be badly designed. They can be designed for other mediums. They can lack empathy. They can lack a strong narrative. Online storytelling platforms should offer insights about their constituent stories that would be opaque to other mediums.

Outside of your own work, are there good examples of the kind of insight you're talking about—either online or off?

No good ones. This is a quality that has yet to be realized, in my own work or elsewhere.

What's your own design process then? In particular, how do you start?

On paper. I do all my thinking on paper. I can't think behind a computer screen. When I sit at my desk with my hand on the mouse, my creativity stalls, and I turn into a robot, efficiently executing what's already in my mind.

It's also important to be willing to throw away everything at any moment, if a better idea comes along. Many of my well known designs (10x10, We Feel Fine, I Want You To Want Me) assumed their final design states very late in the process, often abandoning weeks or months of work. Nothing can be sacred. Better ideas must always be allowed to subvert lesser ones.

Figure 5: Some sketches for the “I Want You To Want Me” project, which is on view at MoMA.

I've heard you do all your work yourself, so I suppose you don't need to worry—at least as much—about throwing everything away. How does working alone, rather than hiring a team, benefit you or your work?

I believe in practicing one's own craft. You learn an astounding amount from the resistance of the medium. My projects always evolve enormously from initial concept to final form. Often you try something and it sucks, so you go in another direction. Often you make a mistake, and the mistake ends up showing you a better way. But unless you have an intimate relationship with the medium and with the piece, you will not notice these cues, and the work will suffer.

I think this is why so much of the work that comes from large companies is so mediocre. It gets specified by someone and executed by someone else, with feedback from someone else and final say from someone else. It's just a big mediocre mess. To make something really beautiful, you have to treat it like a lover. It has to be personal. It has to obsess you when you're falling asleep. It has to be in your dreams. It has to be with you when you wake. It has to torment you.

If you allow the work to accompany you in those intimate moments, it will reveal itself to you, and the result will evolve like a life form, nuanced, and crafted with love. This is why I resist hiring assistants or interns, and certainly why I will never work for anyone else. This resistance to hired help limits my ability to produce more, but I think it makes each thing I do produce better.

Figure 6: A screen capture from the “I Want You To Want Me” project.

Can design studios or even designers in large companies foster this same sort of obsession?

No, I don't think so—not with the same sense of loving. The best stuff is always personal.

It seems that your work is indeed becoming more personal. In earlier projects, you created tools that reveal storylines by analyzing large sets of data—but you yourself were removed from the actual story narrator. In more recent projects, you're actively involved in the story, becoming the narrator yourself. How has this shift affected you as a designer? Has it affected the way that story gets told?

This transition is partially selfish (I don't like the person I become when I sit for weeks behind a screen), and partially conceptual (I think the most moving work is personal work). After spending five years making projects that passively collect and reframe the stories of others, I began to see in this work something superficial, even cowardly. Piggybacking on all these other people's stories, without risking anything myself, began to get to me. So I'm heading in a different direction now. And we'll see where that leads.

You've been revealing narratives from the time you were at Princeton when you founded Troubadour Magazine through your work today in the MoMA. Have you seen a shift in the kinds of narratives we find interesting? Well, one big change is people's obsession with data, which didn't exist seven years ago when I was at Princeton making Troubadour. People have come to expect to see the story of data alongside the story of people.

In fact, I think people have begun to forget how powerful human stories are, exchanging their sense of empathy for a fetishistic fascination with data, networks, patterns, and total information. Really, the data is just part of the story. The human stuff is the main stuff, and the data should enrich it.

Isn't this contrary the trends that seem to be emerging today with increasingly sophisticated technology?

One thing technology hasn't figured out is how to enhance, rather than limit, our individuality. This is largely because technology, in order to be profitable, has to be mass-produced. And when things are mass-produced, everyone gets the same version. This would all be fine, if technology weren't so damn addictive.

Most tools (hammers, pencils, leaf-blowers) don't impose themselves on the lives of their users. Tools should accommodate us when we need them, and then go away until we need them again. This is the implicit pact between tools and their owners. But technology is different. Devices like Blackberries, iPods, iPhones, and laptops are the first tools to violate the pact of tools and their owners. They go too far, and they take over our lives.

I don't have anything against these devices in particular; I just think every so often, people should go outside and smell the wind.

How do you judge where that line is so that you don't violate the pact of tools? Is there a way designers can better understand when tools should retreat to the background?

The problem is less with the tools and more with people's addictive relationships to them. But if tools insist on becoming addictive, they can at least have the courtesy of being personal. Instead of one-size-fits-all solutions, objects and experiences could fit the lives of their owners like a tattered favorite t-shirt, a lucky necklace, a haircut, or a snapshot of a first love.

Designers tend towards fascism—creating products and experiences that force themselves onto their users, leading to addictive relationships that all look the same. If these objects of technology are going to be part of our lives at all times, then they should bend and morph to reflect our personalities, enhancing rather than diminishing our sense of individuality.

Which story most influenced you as you were growing up?

I had this great big book called the Encyclopedia of Things that Never Were, which was a compendium of all things magical, mystical, and mythical. I also read a lot of comics (and wrote my own).

In a recent interview you said, "There are many ways to tell stories, and I'm still searching for the ultimate way. I don't think humans have found it yet." How do you push yourself to find new ways to expose these narratives?

It comes from living life and having an extremely intimate sense of technology. These things play off each other, and you start to see the physical world through the lens of technology. You can then envision applying templates and frameworks on top of reality, to highlight and reveal insights and patterns about life. It's an organic process that only comes with time. As Hemingway often said, you have to live life to make work.

How do you find stories to tell?

Oh, they're everywhere! Everything is a story. Sometimes it overwhelms me. I love walking around New York and making intense eye contact with strangers as they pass me, sharing a few brief moments of knowing, guessing intimacy.

Every face holds so much. I could do an entire project about anyone I ever meet. I just find people endlessly fascinating.

Where to go from here

If you enjoyed this article, check out these other great Think Tank articles:

• Graduate education: Preparing designers for jobs that don't exist (yet), by Anne Burdick
• On the ground running: Lessons from experience design, by Adam Greenfieldk

About the author