Ben Fry
Reimagining How We See Data

---

When scientists complete a huge undertaking, such as the Human Genome Project, the public often has no clear picture of what they have accomplished. Sure, the effort was referred to as "mapping the genome," but no map or periodic table-like image exists to help interpret the massive amounts of data contained in our genetic code.

Enter Ben Fry, who runs a design and software consultancy in Cambridge, Massachusetts. His goal: Putting a picture to the data in order to better convey its meaning. Fry's work is based on his philosophy that "the visual side of manipulating information is inseparable from the utilitarian side of it."

"I use an enormous amount of UNIX-ware. Mac's combination of having the UNIX apps, but also the Adobe apps, like Photoshop and Illustrator, allows me to get things done."

Utilizing his background in graphic design and computer science, Fry fights the compartmentalization of data handling—the separation of the statistical number crunching from the interpretation—while adding new depth to what can be conveyed through design. His work has appeared in movies, such as Minority Report and The Hulk; in New York City museums, such as the Whitney Museum, the Museum of Modern Art (MOMA) and the Cooper-Hewitt Design Museum; and in print publications, such as SEED and on the front cover of the journal Nature.

From Data to Understanding

Fry began his visualization work in 1998, as a Ph.D. candidate at the Massachusetts Institute of Technology's Media Lab. He has tackled several different data streams, from the text of Mark Twain's travel memoir The Innocents Abroad to links between popular blogs to U.S. zip codes. Once Fry identifies a question he wants to answer within an information set, he goes through a process of transforming raw data to what he calls "understanding." The process requires merging several fields: computer science, statistics, graphic design and human/computer interaction.

Fry works most often with the human genome, a draft of which was made publicly available just as he began his doctorate. He characterizes the genome as "an interesting, difficult, messy, complex problem." Genetics, he says, is a subject that laypeople typically only hear about through the filter of the media or maybe from science fiction movies, like Gattaca.

Thus far, he's tackled a number of different genetic questions to try to make the subject more tractable: How can we depict the evolution that took place between humans and several other mammals, including elephants, dogs, cows and opossums? How different is our FOXP2 gene—which is credited with giving us our ability to have language—than that of our closest relative, the chimp? How can we show differences between the genomes of inhabitants from different regions?

Mac as a Base of Operations

All of Fry's pieces are developed primarily with Mac—specifically a MacBook Pro outfitted with a 30-inch Apple HD Cinema Display. His projects need to work across platforms, however, so he also keeps some Linux and Windows machines around. "Because the Mac plays nice with Windows and Linux—more so then they play with each other—it's a nicer sort of base of operations," he says.

"I don't have to turn over to a UNIX workstation to do UNIX-type stuff and switch over to my other machine that makes pretty pictures. Since the crux of my work is that you can't really separate those two, Mac makes it an ideal way jump to be able back and forth like that."

Fry typically begins crunching data with Perl or another scripting language, as well as other UNIX-based software, such as the text editor Emacs and the compiler GCC. Along the way, tests are sometimes performed with the help of Adobe Photoshop or Illustrator, before finalized pieces are completed in software. Mac OS X, he notes, can flawlessly handle both the unglamorous analytical work, as well as the sexier design elements. "I don't have to turn over to a UNIX workstation to do UNIX-type stuff and switch over to another machine here that makes pretty pictures," he says, "Since the crux of my work is that you can't really separate those two processes, Mac is an ideal way jump to be able back and forth like that on one computer."

Processing with the Help of Mac

The most important tool in Fry's software arsenal is Processing, a Java-based programming language he developed with collaborator Casey Reas. It's free and open source and is an environment that allows users to generate images, animations and other visual media. While Adobe has recently added some Javascripting capabilities to Photoshop and Illustrator, the amount of information Fry uses in his projects would crash those graphics programs.

Processing simplifies all that. In a project called "zipdecode," where Fry tried to determine how the zip code-numbering system works, he was able to write a single page of code that translated the latitude and longitude coordinates of post offices into a dotted map of the U.S. (As a bonus, the map also ended up portraying how the population is distributed around the country.)

When updating his software, Fry needs to test it on Linux-based workstations and Windows to make sure it runs on each operating system. "My apartment at various stages has been a den of testing equipment to try make sure that these are working on all different platforms," he explains. Now, however, he only needs his Mac: "As VMware has improved, and since the switch to Intel components, I can now run more of that testing on a single machine--so, I can do that on the road or on the plane, which is huge just in terms of the amount of traveling that I do."

New Tools for Interfacing with Data

Fry sees his visionary work evolving along with Mac and its various product lines. He expects to be adding a MacPro to his hardware arsenal in the near future, which will help him attack bigger data sets, such as public data from the ongoing 1000 Genomes Project that will sequence a thousand people's genes to find out more about human variation.