Response to Crisis in a Networked World
View PDF
“It is our hope that emergency management can directly benefit from cyber-enabled analytic approaches, not only in terms of the investigative findings that address changing sociotechnical behaviors in such situations, but also because these same integrations and visualizations support the kind of real-time analytical activity that both members of the public and emergency managers already perform. Such a human-centered computing approach is at the foundation for reframing the information dissemination activities of emergency response as a collectively intelligent activity” -Crisis in a Networked World: Features of Computer-Mediated Communication in the April 16, 2007, Virginia Tech Event
This conclusion is interesting and provocative. While this paper focuses on the social media artifacts after the event, I am compelled to think of how social media could be used to prevent such calamities and minimize loss during such events.
In 2009, an MIT project entitled Gaydar rattled the media. The student developed program crawled social networking sites and was able to determine the sexual preference of a user with an 85% accuracy. The software “examines what the connections between people can tell us, from predicting who might be a terrorist to the likelihood a person is happy, fat, liberal, or conservative. “ - http://yro.slashdot.org/story/09/09/20/1753254/MIT-Project-Gaydar-Shakes-Privacy-Assumpitons Given the accuracy of such preliminary algorithms, this seems to be an area of study in need of additional investment. Perhaps algorithms could have determined that Seung-Hui Cho was severely disturbed, flagged him and notified the school... After all, he had sufficiently archived his troubled mind on the internet : 
Facebook already uses facial recognition software to tag photos, and Google's first for-profit technology deals with facial recognition. Why not dispatch these applications to identify fugitives and missing persons? Clearly, implementation of this type of social data mining and social profiling is highly controversial. In my opinion, it should not be. Users of the internet need to understand that once information and media hits the cloud it is public. As Hal Abelson, the advising professor for project Gaydar was quoted : "That pulls the rug out from a whole policy and technology perspective that the point is to give you control over your information — because you don't have control over your information."
Another pre-implementation of social media in disaster scenarios would be the integration of alert systems with first responders. In my opinion, there is no reason why Facebook and Twitter at least should be integrated with 911 systems. There was an incident in Aug 2010 where a woman typed an IM conversation on her laptop with her toes after being tied up by an intruder http://www.cbsnews.com/8301-504083_162-20012683-504083.html . How about a Twitter hashtag #911 that pickups up your IP address for location, or GPS coordinates from your phone and dispatches help. Perhaps the most obvious implementation of social media for disaster response and prevention would be the integration of the EAS. When there is a tornado, flood, warning – why not SMS all mobile users in the area?
After hearing your story about the 911 text in class, I really starting thinking of ways that we can find a solution to that. I haven't called 911 recently, thank goodness, but I know that the last time I called, I could not be tracked because I was using a cell phone. I was in a car accident and in the middle of an intersection. I couldn't tell them where I was right away because I wasn't sure. They needed an address. There has to be a way that we can pick up IP addresses with Twitter and 911 hashtags. If a hacker can pick up my IP address, why can't emergency responders? We all have GPS on our phones than can immediately connect to google, but can it be accessed by others? (I'm asking because I feel like you might actually know the answer.)
ReplyDeleteSMS to all people in a selected area might be difficult. If you went by area code, you have to remember that almost every single person now has a cell phone. Unlike before, long distance calls are not an issue and people almost never change their numbers anymore. Do you freak people out by informing them of a tornado though they now live in Alaska? If it was a sign-up type of thing, would you send notices to every single home in the state? That would be an incredibly expensive project and wouldn't promise results. What about when phone lines are down because of tornadoes nearby? Or if Xcel is fixing something that day-- does that neighborhood get the warning? Would it matter if people had different service providers? Just a few thoughts! Not saying that it isn't a great idea, I think it is. I'm just curious as to how you'd reach the broadest audience in time.