For the Greater Good
During a large-scale disaster relief IT project, Chamindra de Silva and his team found that managing in times of crises involves quick thinking and flexibility.
PHOTO BY DOMINIC SANSONI
One of the most difficult things about responding to a natural disaster is the sense of urgency. People need things yesterday, so project teams must function in a crucible of chaos and intense pressure.
Such was the case with the Sahana project, which brought together a team of volunteer computer engineers to build a disaster management system in the wake of the 2004 Indian Ocean tsunami. The team was organized by the Sri Lanka-based not-for-profit Lanka Software Foundation (LSF), which is dedicated to promoting free and open-source software. The first functional pieces of software were built in a matter of days.
“It was quite tough, as the deadlines were already past when we started,” says Chamindra de Silva, one of the chief project managers of the project and currently technology strategist in the Global Technology Office of IT outsourcing company Virtusa in Sri Lanka. “A lot of us worked overnight, losing sleep. Building a disaster management solution at the point of a disaster is certainly not the [ideal situation].”
The project arose out of a lack of readily available, up-to-date disaster management solutions to handle large-scale disasters. “We had reached out to many international agencies to ask for such solutions, but none were available, or the ones that existed were based on legacy technology or were outdated,” Mr. de Silva says.
Initially, hundreds of developer volunteers were given time off by their respective employers to work on the project. But eventually, the core group tapered down to about 20 people. “In the early days, we had a 24/7 operation, and the first bits of software went into production use in about a week,” Mr. de Silva says. “Over time, more and more capabilities were added and used in various ways.”
During the first phase of the project—and more strictly in the second phase—the group used a methodology called extreme programming to churn out software quickly. Extreme programmers break projects into small chunks, develop simple code, test often for quality and quickly deliver functional bits of software as they are ready, before immediately moving on to another section.
Mr. de Silva learned that when it comes to managing during a crisis, flexibility is best. For the Sahana project, the team created a modular architecture that allowed it to break into smaller groups of three or four to work on various aspects of the system. “The requirements for each module had a different set of stakeholders, which enabled us to work independently,” he says. “In the instance of a disaster, you find that strict chains of command create bottlenecks.”
But disaster management is about more than the reactive. The group recognized the need for a readily available disaster management system worldwide. So LSF collected enough funding to support a revision of the system—the free and open-source software is available for download at sourceforge.net/projects/sahana. The system has since been deployed in disasters such as the 2005 earthquake in Pakistan, the 2006 mudslide in the Philippines and the 2006 earthquake in Indonesia.
Leadership 2008 / www.pmi.org