Early last month the Department of Defense decided to put up party decorations across the country in commemoration of the Internet’s 40th birthday. On Dec.5, the Defense Advanced Research Projects Agency (DARPA) distributed ten 8-foot red weather balloons across the continental U.S. and challenged teams to be the first to correctly locate all ten.
Since the balloons’ locations spanned from San Francisco to Miami, the challenge was not intended to be solved by a team of incredibly ambitious travelers. Rather, the goal of the challenge was to locate the balloons by managing a large social network and tracking spreading data online.
According to the challenge’s , the challenge was meant to, “explore the roles the Internet and social networking play in the timely communication, wide-area team-building, and urgent mobilization required to solve broad-scope, time-critical problems.”
The Tech team, nicknamed “I Spy A Red Balloon,” was headed by Erica Briscoe, a research scientist at GTRI, and Ethan Trewhitt, a research engineer at Georgia Tech Research Institute (GTRI). The team also consisted of Dr. Betty Whitaker, Stephen Cuzzort, Jessica Pater, Rick Presley and Miles Thompson.
When asked why they decided to participate, Trewhitt said, “For the most part, it was for the fun of it. I think it played more to our general analytical interest [than our specific research].”
Tech’s team spotted nine of the ten balloons by the end of the day, coming in second overall—just behind MIT’s team. The MIT team found the tenth balloon almost seven minutes before the GTRI team. However, the GTRI team found nine balloons, thus making them the singular team to find that nine balloons by 6:59:11 p.m. The third place team found only eight balloons at a time of 6:52:54.
Though the team didn’t take home the $40,000 top prize, they made their decision with what to do with the prize money before the contest had even begun.
Trewhitt said, “We started by making a conscious decision to donate our money to charity.”
According to Trewhitt, few teams planned on keeping the winnings for themselves.
Trewhitt said, “There were two schools of thought at the start: either donate the money to charity or distribute it to people who found the balloons.”
The team decided against the latter in order to avoid the tax headaches that would come with it and avoid the unfairness of someone finding a balloon on chance while other team members spent hours looking without success.
Instead, they decided to donate any winnings to the Red Cross, a decision that won them several supporters in the challenge.
According to Briscoe, the team spent most of their time the day of the event watching their social networks rather than actively searching for the balloons. Briscoe said, “We heard there may be an Atlanta balloon, so we sent some people to find it, but also monitored [our sites].”
The team drew a large number of supporters before the event was over. According to Briscoe, the team’s had 600-700 people sign up to watch for balloons, and the Facebook group gathered another 800-900 supporters.
However, part of the contest wasn’t just about gathering information; according to Trewhitt, several teams and individuals outside the contest succeeded in throwing teams off the trail.
Trewhitt said, “There was definitely a counter-intelligence effort. A lot of misinformation popped up on Twitter and Facebook.”
Briscoe said, “Twitter in particular was bad. Things on Twitter spread like wildfire whether they’re true or not.”
Many of these simply proved to be minor distractions, though, as the team would contact people in the area to go out and check sightings for themselves.
Once one team realized the ruse, the word got out on Twitter just as quickly as the misinformation.
The event was partly aimed at just that: tracking how fast information and misinformation can spread in the hyper-connected modern world.
Between social networks and new ways of obtaining and bargaining for information online, DARPA wanted to see what it could learn about the spread of information in online age.
Trewhitt said that DARPA also meant for the event to act as a public relations outreach to researchers across the country, so the entire affair had a fairly light-hearted, fun air about it. The red balloons replaced the original targets—ten DARPA employees—for just this reason, as some were concerned what government-sponsored tracking of individuals would look like.
Trewhitt and Briscoe both say that one thing was left unresolved at the end of the contest: the identity of the person who reported the location of the Atlanta balloon.
The team located the balloon through a tip on their website, though the only information they have on the tipster is the name “Emma” left on the site. Despite the submission being made from a Tech connection, the team has been unable to thank the tipster.