To most Americans, Paul Revere was a hero. But what would the British have learned about him back in the 1770s if they employed the type of spying techniques the modern-day National Security Association has been accused of using?
Kieran Healy, a sociology professor at Duke University, explored this question in a recent blog post written in the voice of a British intelligence analyst.
Healy wanted to show how metadata, like the phone call logs that the NSA surveillance program of today relies on, could be used to identify “key persons involved in terrorist groups operating within the Colonies.”
And the approach was very effective.
Revere popped out at the center of Healy’s network diagram: 254 dots showing colonial Bostonians connected by a tangle of lines marking their common social group memberships.
And though the results weren’t new or unexpected, the presentation of the data made the work resonate.
Healy’s post had attracted more than 260,000 views in less than six weeks, about three times his typical blog traffic.
Healy is one of a group of Triangle-area researchers who are working to create easy-to-understand graphics to share complicated concepts with broader audiences. Like YouTube videos of water-skiing squirrels or laughing babies, some scientific online creations get passed around social media.
“There’s a strong incentive to work to produce graphics that are compelling and honest and relevant that people will want to share,” he said.
Healy says it’s hard to make matrix algebra, one of the math techniques in his post, entertaining. But he managed to keep the audience engaged with his old-style English diction, playful tone and convincing graphics.
“If you can be funny about it in an engaging way then that really helps,” Healy said.
Colorful language maps
Josh Katz, a Ph.D. student in N.C. State University’s statistics department, also works hard to make data easy to understand by producing colorful maps showing word usage across the United States.
Less than 48 hours after his maps were posted on NCSU’s webpage in June, Matt Lauer was pointing to them on a screen on the “Today” show, Katz said.
But just like in Healy’s Paul Revere blog post, the information in Katz’s maps wasn’t new.
The data, tens of thousands of responses to 122 survey questions, were originally presented as individual colored dots by Bert Vaux, then a professor at Harvard University.
But Katz felt the dots were hard to interpret. After running statistical analyses to create nationwide gradients of the language data, he worked to produce maps with visually accessible color gradients.
And the public could easily observe how the nation pronounces the word “pecan” in different regions, or which parts of the country use the word “soda” to describe a carbonated beverages, and which ones use “pop.”
“The one thing I still can’t get over is how quickly the site just caught fire,” Katz said.
Touching the nano-world
Russell Taylor, a research professor at the University of North Carolina-Chapel Hill, has developed a way for scientists to touch and interact with objects too small to see, such as the fibers making up a blood clot. Using virtual reality technology, Taylor creates a floating three-dimensional surface 100,000 times larger than the original object. Scientists can feel surfaces of the projection using a special pen that moves and vibrates.
Just like Katz who experimented with fitting data onto a language map, Taylor has worked to effectively display information on screens and surfaces.
One method uses stacked dots representing information like temperature and humidity. Successively smaller dots are layered on top of bigger ones that remain visible underneath. By using the dot approach, however, some information is lost.
“There’s a tradeoff about showing something clearly and showing something that is correct,” Taylor said.
And whether it’s Healy writing on Paul Revere, Katz’s dialect maps, or Taylor’s 3D displays, each data representation must find the right amount of information to show in order to balance clarity with completeness.
Katz is just happy that people outside his academic field have been looking at his maps.
“It’s good exposure for statistics, for thinking about data in creative ways,” he said. “It’s just been a win.”
Before experiencing Taylor’s interactive three-dimensional display, scientists are sometimes skeptical of its benefit. But after they see and feel their own data, they often learn something new about the tiny object they’ve studied for years.
Taylor described a typical reaction to a scientist’s first interaction with a 3D projection: “Wow. Can you make it do this? Can you make it do that?”
In response to his Paul Revere post, Healy received lots of email.
“It’s remarkable how a compelling visual can spread really fast,” he said.
And readers really connected with the Paul Revere post. Some were surprised they liked the math that Healy presented. And one message of praise even came from a descendant of Paul Revere.