In just a few hours of coding, Jonathan Mendez and Greg Stoll were able to create what Wired News called A Disaster Map ‘Wiki’ for Hurricane Katrina. Jonathan was able to share his thoughts about their Katrina Information Map project in an e-mail interview with us yesterday.
Q: What gave you the idea to the Katrina Information Map?
The idea for the map really sprang from my own need for information. My family evacuated to my home before the storm hit, and we then spent a lot of time just trying to find any information that we could about how their house was affected and about how friends had fared. Information from the news stations was focused, appropriately, on the most drastic aspects of the tragedy, and weren’t much help in finding out about our neighborhood. On the Internet I found forums set up for people in different areas to give or ask for information, but they of course had hundreds of posts, filled primarily with questions with very few answers, and were hard to comb through. While browsing the forums I thought that it would be wonderful if there were a map telling us the status of different areas, and since I doubted the authorities had the time to go street by street to record this sort of thing, especially with how hard and dangerous navigation has been there, I thought that the information would have to come from these personal accounts that people had. At that point I thought of the Google Maps API, and I knew that my friend Greg had created a map with a static set of markers, so I figured it wouldn’t be a big leap to allow people visiting a website to add those markers themselves.
Q: What has the response been like? How many postings? How many people have told you it was helpful?
The response has been tremendous, far more than I originally expected. I posted the link on the few forums that I knew of, and saw the initial map start to slowly gather a few sparse entries, and thought maybe people didn’t have time to go to the map to add information. Within a few hours, though, it really picked up and I saw probably close to a hundred markers already. Checking the access logs for the web site, I saw more and more sites linking to the page, and by now the map has 1,136 entries, with more being added all the time. I’ve also gotten a lot of feedback praising the site and in many cases thanking me for creating a site that allowed them to find out information about their neighborhood or their families and friends. It always brings a smile to my face when I get an e-mail in which someone describes finding out good news about their loved ones.
Q: It seems like with a few improvements, this application could find a lot of uses. Knowing what you know now, what improvements would you make?
I’m sure that it could definitely find many uses. I’ve definitely realized that I should keep the code around in case, God forbid, another widespread tragedy would find use for it. And we’ve definitely thought of several improvements that could be made, some of which we hope to still make while this site is being used. One comparison that has been drawn many times has been to wiki, the format used most famously by Wikipedia to allow anyone to add/edit entries to a website. I think that the site could benefit from some of the features of a true wiki, such as being able to edit the markers (right now, one can only append information to existing markers, or add new ones), and most importantly, allowing people to see the history of a marker and also to register for notification of updates to a marker, allowing community policing of the site. Another idea that has been suggested to me that I think would be wonderful is enabling in-the-field submissions via SMS, or some other similar system. This would allow the application to be run by the teams deployed to the area, enabling them to put up official information straight from the location as they move into the area. The map could be supplemented by photos taken from the area, and so forth. There are many improvements that could be made to make it more powerful and more informative to the people using it.
Q: The Associate Press put together a similar and impressive map, but it did not go the extra step and add collaboration. Why do you think that is? What are some of the encouraging or discouraging things you’ve learned about blind collaboration?
I think that allowing blind collaboration on a site run by an organization that is held to high standards by the public would have to be done very carefully to prevent hooligans from discrediting the site or spreading misinformation. That’s one of the bigger problems we’ve had to face on our web site, as people have been posting obscenities or other entries they consider to be clever. For now we’ve had to manually remove such entries by monitoring each new one that is made. Of course, this method only protects against the nonsensical entries; unfortunately we have no way of preventing someone from posting false information that appears to be realistic. That’s one of the problems any collaborative effort has to face; I’ve already talked about some of the ways to combat that, but we don’t have the time to develop such methods into our site during the timeframe in which it is still useful. Of course, the great benefit of the collaboration is the wealth of information that is quickly added to the site by the random people who know something. Unlike a formal news organization that must have some significant portion of information to be able to spread it (I’ve never seen the news say “The third house from the dead end on this street is mildly damaged from the wind but was spared flood damage” for example), collaboration allows these small pieces of information to be used, and then they get put together to form the bigger picture.