Emergency Response Symposium

This morning I attended the Emergency Response Symposium at Dalhousie University. The symposium addressed "emergency response to disasters (e.g. earthquakes, fires, floods, ice storms, epidemics) that require coordination of diverse resources distributed in time and space, where decisions have to made with limited information." Dr. Mark Elmore of Oak Ridge National Laboratory was there presenting.

The first thing I noticed, was that the room was split between university academics (professors, Ph.D students) and actual disaster response people (Halifax Fire Response, RCMP, Joint Task Force). I got the impression there was a total disconnect between what the two groups were saying. The academics were talking pure theory (graph, data, matrices, analysis) and the disaster guys were talking about reality (gut feeling, life, death, response time). Let's just say, there is a lot to be said for applied-knowledge vs pure-theory. Here are some of the key points I took away:

• A really fast search doesn't help you, if it gives you the wrong results. When response time is so critical, having the wrong data at the wrong time is matter of life and death.
• Past data can be used to make better decisions in future, but how do you capture 'gut feeling' of a fire fighter that prevents him from going into a room on fire? How can this knowledge be applied to an algorithm for better decision making? Or does this decision need to be left in the hands of the user?
  
• A big challenge is getting past information like building layouts and maps, combined that with real-time data like weather and position of response teams, and render in 'no user manual required' interface to allow fast decision making. Response teams don't have time to analyze charts or graphs.
  
• There is a major problem with data lost when someone retires or quits. Imagine a firefighter retiring after 30 years, all that field experience is lost. How can it be captured? How can it be re-applied to make better decisions in the future?
• How can crowd data be verified? In mission critical data, someone needs to be accountable when lives are lost, can we hold the crowd liable?
  
• Public vs Private. There are cases like fires, where public data of the crowd can be used to help make better decisions. One recent example was the Halifax Fire, where Twitter was on fire with activity. But there are private cases like with the military, where public data is not accessible. How can similar decisions be made without access to this public data?