One year during a keynote address at a PASS
conference, Bill Baker (GM for SQL Server Business Intelligence) talked about the "goodness of databases." The idea appealed to me; as DBAs a big part of our job is to do our best to guarantee the accuracy of the data we manage. If the data is accurate, it's useful and it helps people do their jobs, or manage their lives.
The goodness of databases was magnified a thousand times in the aftermath of Hurricane Katrina. Accurate and available information became crucial to allowing families to be reunited. In the initial days after the storm, many sites were used to connect people. Existing sites (craigslist, yahoo boards) were utilized, and seemingly every news organization hosted a message board for people searching for loved ones.
There was lots of information, but it was decentralized. An evacuee in the Astrodome might have 20 sites to visit to look for a lost spouse or sibling. Seeing a need for coordination and centralization, two projects emerged: Katrinalist.net
. Katrinalist is an open source project run by volunteers, managed using a wiki
and listserves. Its goal was to aggregate the data from dozens of sources into a central repository. There were two main subprojects; one was to design and publish a standard for data interchange to allow bulk data loads. The other was to scrape sites prepare chunks of data that could be hand entered into the katrinalist database. A huge number of volunteers stepped up to do the data entry. The Katrinasafe project was a joint effort between Microsoft in collaboration with the Red Cross to centralize all the information about evacuees collected at Red Cross shelters.
There were other efforts to centralize information, and my point is not to do an exhaustive review of all the efforts. But they hall have one thing in common: they illustrate the "goodness of databases". The availability of accurate information was critical to thousands of people trying to put their families back together.