Another impromptu interview this time with Dr. Asim Ijaz Khwaja, economist and assistant professor at the Kennedy School of Government on his work with RISEPAK (Relief and Information Systems for Earthquakes in Pakistan). RISEPAK is a web 2.0 tool that was launched after the devastating 2005 earthquake in Pakistan to aid in 1) the rapid dissemination of information and 2) real-time coordination between typical relief actors (e.g. Red Cross, Mercy Corps, etc.) and non-traditional actors (non-relief NGOs and people in the field who were trying to help).
Duration: 1min 53sec
“By and large”, Khwaja told the assembled group at the 2008 Harvard Social Enterprise Conference, “the people who die in natural disasters do so simply because we can’t reach them quickly enough.” By quickly enough, he means within the first few hours or days after the disaster strikes. He goes on to say “In the 2004 tsunami in Sri Lanka, 61% of the survivors received no help in the first 2 days… In the 2006 South Asian earthquake in Pakistan, there were thousands who didn’t receive anything in the first week. Why is that?”
Dr. Khwaja and the rest of the RISEPAK team identified 4 possible reasons that they thought their site could address:
- Information. Real time (up to the minute/hour) data on where, what and how much relief is needed doesn’t really exist a disaster. The potential actors are often flying blind and relying on educated guesses.
- Coordination. In the Pakistan earthquake, there were many non-traditional actors (e.g. NGOs in the field who had never done relief work) who wanted to help, but had no way of coordinating with the centralized disaster response efforts. Many would simply load up vans with supplies and take them to where they thought they were needed most or where they could get to. In fact, a substantial fraction of the initial relief was done by these non-relief NGOs or individuals. Another [unsurprising] observation of these uncoordinated efforts was that the easiest people to reach (e.g. communities along the main road) tended to receive lots of aid, while harder to reach communities, say across the valley, tended to receive nothing or far less. A big question is how to coordinate these actors?
- Aggregation. The RISEPAK team noted that while the larger relief players knew the granular data of how they were impacting the relief efforts, (i.e. the number of trucks they had deployed, the amount of supplies sent, and to what districts), finer data about exactly which villages in the district or which people in a village was harder to come by/did not exist.
- Accountability. During a disaster response, many tend to view accountability as something that occurs months into it. The RISEPAK team found that real-time accountability was particularly necessary.
RISEPAK provides village level geographic, demographic, and satellite data of the affected region that include maps, census data/number of people, roads & accessways, elevation, etc. People send in information on current access, damage, what has been delivered and what is needed via fax, phone, text, and online form. Within a 8 hour turnaround, this info is collated and posted on the site. Dr. Kwaja acknowledges that there is a chance that people will try to game the system, but hopes that this will be less likely in the initial stages of disaster responses.
His ultimate dream is for a world-wide system of this kind rather than just for Pakistan. This would be a great nomination for a TED Prize.
Other uses of web 2.0 in disaster/humanitarian response