5 Lessons From the Tech Response to Haiti

Nearly two months ago today, Haiti was devastated by a 7.0 earthquake, and the tech community went into overdrive to try and make the relief effort smarter and more responsive. Looking back, what are the most important lessons we can begin to draw, and how should we think about how our systems can evolve to meet future disasters?

Lesson #1 - Information is king: The tech response to Haiti was all about information. It was about finding better ways to collect information from people actually on the ground, process that information and visualize and distribute it to people who needed it. This focus on data is representative of a larger, more general emphasis today on the fact that more data and better aggregation means similarly better ways to make decisions.

Lesson #2 - No one has the whole answer: One of the most heartening aspects of the tech response was the way different actors quickly figured out their specific roles and linked up to make it happen. Ushahidi had the actual mapping of information covered. InSTEDD was on the ground to help coordinate information use and spread the word on how people could be reporting emergencies. FrontlineSMS took charge getting the "4636" short code established to make it easier and faster for people to report information. As the volume of information increased, Crowdflower retrofitted its microtask system to help volunteers translate and add geographic information to the messages and reports pouring in. And microwork company Samasource retrofitted a Haiti project to actually get locals involved translating the messages.

Lesson #3 - Wide networks pay off when the time comes to roll: Part of the reason  groups were able to work together so effectively is that they all were talking to each other in advance. This is a community of organizations that actually works pretty diligently to know what the others are up to. That communication paid off in a serious way when it came time to move quickly.

Lesson #4 - People want to help, so give them a chance: The response required grabbing data from a huge number of people, using a likewise huge group of people to make sense of that data and then again distributing it to a crowd. This was crowdsourcing at its most robust. And one of its lessons, articulated well by Crowdflower CEO Lukas Biewald in this piece, is that using distributed work forces is often smarter and more cost-effective than trying to establish teams that require training and maintenance.

Lesson #5 - Thinking short-term doesn't mean not thinking long-term: Finally, for most of these groups, their short-term actions had long-term implications. Ushahidi's platform is designed not just as a way to visualize geography, but to show patterns over time and preserve them as an artifact to be learned from. Samasource is taking this committment to the long term a step further, and building a work stream from microtasks that can employ some of their local volunteers who want to commit to more than the immediate quake response.

I'm sure that many of the people involved experienced a litany of complications that were too challenging, and which they wished had worked better. But I still think that overall, what we saw in Haiti was an exemplary model of cooperation. Check out this video produced by Crowdflower during the crisis:

Mission 4636 from CrowdFlower on Vimeo.

Photo credit: USAID_IMAGES