Update on DARPA Red Balloons, MIT won

Just in case anyone was wondering, the MIT team found all 10 DARPA red balloons yesterday and did it in 9 hours. Part of this team's approach was similar to other teams - networking on social media sites and suggesting friends tell friends, etc. to find the balloons. But MIT also had this at work for them...

How do you rule out the dishonest reports of spotting the balloons? This is one of the most interesting parts to the challenge! We will use sophisticated algorithms from the field of network science and complex systems theories along with machine learning algorithms to identify valid submissions.

Well, when you're MIT, you've got that kind of know-how and clearly, it was a helpful tool for them. I'm sure the DARPA folks will be very interested in learning more about it when they interview the team about their strategies.

During the day yesterday I was checking in with Twitter via my BlackBerry. L. and I read a couple of reports that there might be a balloon in Portland, Maine, so after G.'s photography show at MeCA, we drove around Portland (and then the Maine Mall) to check things out. It turned out that there was a balloon in "Portland," it just happened to be on the other coast.

We also saw reports of sightings in Rhode Island and Albany, NY, and several other places around the country that turned out to be false. Today I found a map on the DARPA site that correctly identified all the balloon locations, some of which I never saw mentioned on the web at all yesterday.


The team I was part of, I Spy A Red Balloon, was out of Georgia Tech and they came up with eight of the ten balloon locations by the end of last night. Had their been a balloon in the Portland (Maine) area I would have reported it both to I Spy team as well as the 10 Balloonies team (the Geocacher's team). My family has been playing the geocaching game for a few years so of course we were going to share with them too (I wasn't aware they had a team until after we had joined the Georgia Tech folks).

The geocaching team had pledged to donate their winnings (DARPA was giving $40,000 to team for finding all 10 balloons first) to charity as well as an additional $10,000 from Groundspeak (the folks behind geocaching.com). It turns out that even though the 10 Balloonies team didn't win the challenge, they're still donating their $10,000 pledge to charity anyway. I knew there was a reason I like geocaching and the people I've met playing the game.

But it turns out with this challenge that it wasn't the bleeding hearts interested in supporting charity that won. MIT's team split the winnings among their team's members. I am curious if this played a role in putting them over-the-top in collecting all the balloon locations. Capitalist America is this country's bread and butter mentality (as sad as this makes me) and it's possible that piece could have tipped the scales in their team's favor.

As I mentioned in my previous entry, I couldn't help but wonder about the DARPA motivations in hosting this challenge and saw this reported on CNN.com today:

Johanna Jones, a spokeswoman for DARPA, said the hunt was designed in part to give the military new ideas on ways to operate in a range of situations, from natural disasters to combat.

The agency said it plans to meet with teams to discuss their approaches and strategies used to build networks, collect information, and participate in the contest.

I have mixed feelings about this because of course I want our troops to find the best ways to stay safe when faced with potential combat issues and glad they have the DARPA folks to help them figure this out. But it's still gives me a chill to think about it (yes, I need to get with the reality of the world program but still, one can hope to be naive about some of these things).

DARPA published a press release (it's in .pdf format) when it announced MIT as the winner:

"The Challenge has captured the imagination of people around the world, is rich with scientific intrigue, and, we hope, is part of a growing 'renaissance of wonder' throughout the nation," said DARPA director, Dr. Regina E. Dugan.

“DARPA salutes the MIT team for successfully completing this complex task less than 9 hours after balloon launch.”

DARPA announced the Network Challenge to mark the 40th anniversary of the ARPANet, pre-cursor to today’s Internet, to explore how broad-scope problems can be tackled using social networking tools. The Challenge explores basic research issues such as mobilization, collaboration, and trust in diverse social networking constructs and could serve to fuel innovation across a wide spectrum of applications.

DARPA plans to meet with teams to review the approaches and strategies used to build networks, collect information, and participate in the Challenge.

DARPA is the central research and development organization for the Department of Defense (DoD). The Agency manages and directs research and development projects for DoD and pursues research and technology where the risk and payoff are both very high and where success may provide dramatic advances in support of military missions.

“We need a renaissance of wonder. We need to renew, in our hearts and in our souls, the deathless dream, the eternal poetry, the perennial sense that life is miracle and magic.” -- E. Merrill Root

Overall, yesterday was a mixed bag of results. But when looked at from the perspective of large-scale, national connectivity, it was an interesting exercise in the influence of social networking. Yet, it also offered another lesson that viral networking does not always provide accurate information. Sifting through misinformation to find the real stuff requires some really smart people writing some really sophisticated machine learning algorithms. We social networkers play our part but I'm not sure we'll ever be the star players of the team.