Back in February, Google issued a challenge to its users: “Can you make tax data exciting?”
Turns out, you can. (No, seriously.) Head over to datavizchallenge.org/, and you’ll see over 40 projects that, in their way — through video and graphic and text, through tones both cheeky and sober — answer Google’s question with a firm “yes.” The projects, most of them interactive data visualizations, are entries in Google’s tax-related data visualization contest, the Data Viz Challenge, which asked participants to use the tax data gathered by whatwepayfor.com and come up with a creative way to present it — with an award of $5,000 going to the winner. As the search firm put it in a blog post announcing the contest, “We want you to show everyone how data visualization can be a powerful tool for turning information into understanding.”
Today, tax day, Google is announcing the winner of that contest: Anil Kandangath, whose “Where Did My Tax Dollars Go?” illustrates users’ personal contributions to the federal budget through a combination of text and charts and interactive elements that work together to create, believe it or not, a narrative arc. The site, in encouraging a kind of linear movement-through-data on the part of the user, emphasizes both exploration and explanation.
Kandangath was excited — but also surprised — to learn that he’d won data viz contest. Especially since he hadn’t entered with any hopes of winning. His entry, he says, was really about “trying to scratch my own itch.” Kandangath is an engineer by profession, but “my big passion,” he says, is art. (He runs the site The Naked Frame, an image curation site through which he tries to give exposure to other artists.) “I try whenever I can to bring beauty to the way things are presented,” he told me; particularly from the pragmatic perspective of engineering, “it’s easy to forget about beauty.”
He built the project itself using Ruby on Rails; for the images, he used Highcharts, a Javascript graphics library that’s free and, Kandangath notes, “infinitely hackable.” He also brought an engineer’s precision to the aesthetic elements of the narrative graphic. Particularly toward the end, a lot of time went to Photoshop and simply “trying to get the shadows just right.”
Kandangath had been thinking of doing a tax-related dataviz project even before he found out (through Reddit) about Google’s contest; conversations with his friends (one of whom was sure — sure! — that 40 percent of his salary was going to Uncle Sam) had convinced Kandangath that “most people, including myself, don’t know how money is spent” when it comes to taxes. Before the contest, though, he’d been lacking a data set that would allow him to create something that would be meaningful because it would also be accurate. So the WhatWePayFor information — which Google required all contest participants to use in their data viz efforts, and which was available to those participants through an API — “was a godsend,” Kandangath says.
In fact, the entire contest began with WhatWePayFor, says Jenny Ramaswamy, the Marketing Manager for Google’s Data Arts team. The team saw that site, she told me, “and we thought, ‘Wow, this is really useful. Wouldn’t it be great if we could make that information accessible to a wider audience?'” To do that, Google partnered with Eyebeam, a nonprofit art and technology center that has strong ties to the community of developers who specialize in visualization. They formed a a jury of eleven coders and designers who’d be able to discern the striking and educational projects from the merely striking ones.
It took the group over four hours to determine the winner (“and that was just during the jury meeting,” Ramswami notes). But Kandangath’s entry was finally chosen for two broad reasons: accessibility and elegance. Not only was Kandangath’s “winky Washington” interactive, with its narrative progression and use of varying graphics categories, almost intuitively understandable; it also managed to be comprehensive without being overwhelming. It was an example, Ramswami says, of how “you can take all this information and make it very accessible.”
His goal in creating his visualization, Kandangath says, was to present the data in a compelling enough way that people would actually remember it — the stickiness factor that everyone from engineers to academics to journalists is trying to figure out at the same time. Kandangath’s answer? Narrative. “People remember stories,” he notes. They may not recall individual facts; but when you can craft those facts together in a compelling arc, those facts become part of something more than the sum of their parts: a memorable tale. Even when it’s not rendered in text alone. So Kandangath essentially imposed on himself the task of plotting flat information so that it would come to life via…a plot.
The interactive element of the project is crucial; “Where Did My Tax Dollars Go?” isn’t simply a visualization — a graphic — but an environment. One that Kandangath, the project’s author as much as its designer, has both created and controlled for the user. Kandangath, once he had his hands on the WhatWePayFor info, spent a few weeks simply playing with the site’s tax data and trying to make sense of it; he then got to the business of figuring out how to present it — determining categories, figuring out which types of data merited which types of visualization, deciding how to take users on a data-driven journey. And the resulting interactive, he says, “just evolved over time.”
Kandangath is planning to put the project on GitHub (at some point — “right now, I’m not prepared to show how ugly it is!”). Future, crowd-improved incarnations of the project could allow users to visualize tax allocations for different years, or to see state data rather than federal, or to see similar breakdowns for other countries.
Kandangath’s hope is that the kind of context-driven narrative his project represents will empower information and its users. We talk about facts speaking for themselves; in reality, though, information doesn’t mean much unless it’s bolstered by human understanding. “Where Did My Tax Dollars Go?” — and its fellow tax-allocation visualizations — are a reminder of the great things that can happen when information providers think in terms not just of presentation-to-the-user, but of comprehension-of-the-user. He and his friends often talk politics, Kandangath says; but “the problem is that most of the time we’re talking about perception of what is happening without any real data to back it up.” The tax data visualization was part of a broader ideal. “My goal,” he says, “was to make sure that if we talk about economics or politics, we have a place where we can go” — a place not just for obtaining information, but for understanding it. “And then you can form an opinion.”