Over the past two years, millions of words have been spilled in the popular press about the power of social media to organize citizens against repressive regimes. But what if those same tools could be employed by democratic governments mobilizing citizens to achieve positive ends?
A competition the State Department co-sponsored aimed to answer that question in a very specific way. A fictional group of jewel thieves (portrayed by real people) was scattered across five global cities, and a $5,000 bounty was offered to whomever could track them down within 12 hours. The only information the Internet-organized sleuths had to go on was pictures of the “suspects.”
The results were impressive, organizers of the TAG Challenge said. The winning team’s participants managed to find and photograph three of the five people on the crowded streets of Washington, New York and Bratislava, Slovakia. Suspects in Stockholm and London evaded capture.
“The project demonstrates the international reach of social media and its potential for cross-border cooperation,” said project organizer Joshua deLara. “It’s remarkable that a team organized by individuals in the U.S., the U.K and the United Arab Emirates was able to locate an individual in Slovakia in under eight hours based only on a photograph.”
The winning team said it was not yet ready to say how many people participated pending some further statistical analysis, but the total crowd who took part on the side of the team that won amounted to somewhere between 1,000 and 10,000 people.
The idea of publicizing a criminal suspect’s likeness and offering a reward for his capture is centuries old, and it’s the entire premise behind mainstream media productions like America’s Most Wanted.
“But the incentives behind that are very plain. If you happen to spot the bad guy, you might get some money,” said Manuel Cebrian, an MIT and UC San Diego researcher who co-led the winning team.
Cebrian said his team was only able to mobilize the thousands of people it needed to successfully compete in the challenge by setting up a pyramid scheme of sorts, a technique he has used in past federally-sponsored, crowdsourcing challenges. In this case, people who managed to photograph a suspect got a $500 share of the bounty. But they also got $100 if someone they had recruited to the competition found one of the thieves. Smaller incentives were granted for every participant they managed to bring in to the competition.
“For incentives, it gives you the best of both worlds,” Cebrian said. “(The strategy) has all the best of the viral, social media world, but also the motive of greed.” Cebrian said the incentive structure let the “CrowdScanner” team, co-led by researchers at Southampton University in the United Kingdom and the Masdar Institute in United Arab Emirates, quickly set up large networks of spotters in all five cities.
“Over the last five years or so, we’ve seen citizens being empowered by social media, especially in the Arab Spring, to rise up against their governments. I think what the (U.S.) government has realized now is that they can tap into social media too in order to do things that would otherwise be impossible or very expensive. This is a completely different thing.”
The Pentagon in particular has been interested in leveraging the concept of crowdsourcing to solve seemingly-impossible puzzles.
Last year, DARPA, the Pentagon’s advanced research arm, offered prizes to teams that could put shredded documents back together again by distributing the workload to individuals and computers across cyberspace. It also used crowdsourcing to come up with a concept for an advanced combat vehicle.
And the TAG Challenge is, in many ways, the successor to DARPA’s 2009 Red Balloon challenge, where teams were charged with mobilizing online to locate large weather balloons that had been stationed in cities across the globe.
“We were fascinated by the new TAG Challenge for two main reasons. First, a lot of people wondered if the success in the DARPA Red Balloon challenge was a lucky accident. We were determined to see if we can indeed ‘do it again,'” said Dr. Iyad Rahwan, who helped lead the winning team from the Masdar Institute.
“Secondly, we realized that the TAG Challenge is much harder. Instead of locating static, massive weather balloons, you had to locate mobile people mixed within a crowd. We knew that, should we win this, we would manage to raise the bar for rapid mobilization through social media a significant notch higher.”
Cebrain said he and his fellow network scientists have different motivations, though they might help inform the future crowdsourcing efforts of federal agencies. He said despite all of the hoopla about the role of social networking in world-changing episodes like the Arab Spring, little is actually known in the scientific community about whether and how online communities actually were the game-changer against longstanding tyrants.
“We actually know that mass media is much more powerful than social media in spreading information around the world,” he said. “What we need to understand is how much more powerful it is. Is it a hundred times? Is it a thousand times? What we still don’t have is data about exactly how information spreads around the world.”