After Sandy, Let's Stop the Troublemakers in their Tracks
The following is a guest post from Joseph G. Bock, author of The Technology of Nonviolence: Social Media and Violence Prevention. Bock directs graduate studies at the Eck Institute for Global Health at the University of Notre Dame. He has more than 12 years of humanitarian experience in Asia, Africa, the Balkans, and the Middle East.
As we cope with the aftermath of Hurricane Sandy, we need to learn a lesson from disaster-surviving activists in Haiti. They—along with technical experts, volunteers, friends and family members working and living away from the island of Hispaniola—used “crowdsourcing” for early warning of trouble. Rumors of riots were reported through text messages and Tweets, primarily in places where distribution of relief supplies were planned. Peacekeepers used this information to intervene to prevent trouble.
Crowdsourcing in humanitarian disasters involves lots of people (the “crowd”) contributing to a collective effort to map information so that locations of events (such as someone seeing people pick up garden tools and sticks for a riot) or assets (such as locations of where to get food and drinking water) are identified and, ideally, acted upon. In the case of Haiti, a team of technical experts—mainly from Tufts and Stanford—created a digital map using Ushahidi, a crowdsourcing “platform.” Events and assets are put onto the map, providing a picture of what is where, viewable by anyone with access to the internet. Ushahidi ("testimony" in Swahili) was created by journalists who created a way of plotting what was happening amidst the post-election violence in Kenya in 2008. Journalists used it to track corruption of the electoral process and to picture the dispersion of trouble.
After the Haitian earthquake in December 2010, technical experts, led by a PhD student at Tufts, Patrick Meier, pulled together a team of techno-savvy volunteers to map assets and events in Haiti. Even though most people in Haiti did not have access to the Internet, many of them had cell phones that still worked. They sent information via text message or social media (such as Twitter or Facebook). The volunteers posted the information after categorizing it (water, food, medical supplies, potential riot, etc.), verifying it (if need be), and determining the location. A major challenge, however, was the language barrier. Many of the messages were in Creole, the language commonly spoken in Haiti. So the volunteers decided to crowdsource translation from Creole to English. Amazingly, volunteers, mainly from the Haitian Diaspora all over the world, translated messages, on average, in 10 minutes.
As Sandy approached the East coast, I received a message from a friend in Cambridge, Massachusetts, Margy Avery of the MIT Press, who asked for help in using Ushahidi. Her friend, editor and blogger at The Watershed Post in Upstate New York, Lissa Harris, was using a shared Google doc to communicate with her readership about road closings, shelters, places to get help, locations of where relief was needed, and other items.
After hearing from my friend, I reached out to technical experts who could help set it up. In the process of doing so, we found a plethora of digital maps for disaster awareness and response popping up on the internet for various affected communities along the East Coast. After finding out what was already out there, a friend at Notre Dame succeeded in helping the editor and blogger launch her map at www.watershedpost.com/sandy.
As these crowdsourced maps evolve and consolidate, most of their usefulness will be in helping disaster victims know where to go for help and relief workers where to go to save lives and bring resources to those in distress. These maps need to be used, as well, to prevent crime and civil disorder. To do so, three relatively simple things need to be done.
First, categories need to be added for information on looting and conflict (both rumored and witnessed). Most disasters have waves. There is the shock wave, then a pause, then, often, a crime wave, and, (sometimes) a civil unrest wave. We need to put the categories in place now to be on top of the predictable next wave.
Second, law enforcement officials need to keep an eye on the disaster information sites. This can be done by people who scour the maps for new information. Or it can be accomplished using software that sends out alerts to local police, sheriffs, and soldiers of the National Guard (using something like the “Hunchlab” system developed for police departments by Azavea in Philadelphia). Such software uses mathematics to identify patterns of events that portend trouble in the offing.
Third, law enforcement officials need to intervene quickly when they get a warning. Doing so does not usually require a massive show of force. A tangible indication that official eyes are watching can be all that is required.
Using crowdsourcing and working with official responders, we can stop troublemakers before they cause trouble—theft, rape, murder, or provoking a riot. If not completely preventing it, we can at least put a lid on it before it gets worse. We simply need to engage everyone who wants to help the victims of this tragic storm in a collective mapping process, supported by technical experts, plugged-in to law enforcement authorities.
This is about Americans learning from Africans and Haitians. This is global solidarity in cyberspace. This is stopping trouble in its tracks.