We had a bunch of great projects presented at Dublin Hacks. Here are the details on their specs and focus.
Search and Rescue
S&R has the aspect of integrating smartphones with a central system, with volunteers on an open source map. They are creating an Android app to server communications. Users can search for and see where other searchers have been. The API being built has the capacity to input data via phone. The team is currently working on getting data into the formal database, but not all the paths are properly linked yet. Stay tuned for a fully mapped database soon. S&R are planning for the app to be shown to emergency managers and fire department personnel to facilitate a discussion on what can be viably accomplished in an afternoon at a four person sized team in roundtable problem solving. Additional plans include plug-and-play capabilities for myriad emergency scenarios. If the project continues, the team will test run the app on senior emergency officer’s phones. If this app is downloaded and used in other scenarios, please credit back to S&R.
The team ran into some challenges with their geocoding project; Not as many tweets are geocoded as was previously assumed. In terms of numbers, only 1% of tweets are geocoded. This makes extracting meaningful content and data is extremely difficult. Black Hole used Datasift to flesh out twitter data from around Ireland for the past few hours (15k geocoded tweets on the date of pull). The app also has a ton of thought put into visualizations; Colors are hashtags, and provide an easily parsed at-a-glance over a the course of the hours of data pull. Density of tweets is also viewable vuia heat map. Black Hole can also extrapolate normalization of the data trends over time, which provides insight into data black holes. The team is researching and integrating historical data with future plans to test against known events with black holes. Powermap was used in Excel to handle the data. Additional plans include world domination through selling this data, potential to Twitter for the purposes of humanitarian reporting and parsing.
While coordinating digital responders during disaster releif efforts, it’s difficult to figure out who is doing what and where, etc. Part of a solution to this issue is a wiki of pads, with the capability to create collections for individual disasters. The top level page is a pad hosted on one of the following platforms: hacked, mopped, etherpad, etc. The root page will be a static version of the pad content. In terms of updates, link adds inside the content spurs creation of a link tree from the collection page. When clicking edit or collaborate buttons, users are forwarded to currently up and live pads. Every few minutes, these pads are saved as static page. As a point of reference and redundancy, multiple copies of pads exist across a network, but only one is in use at a time. The future vision of the project includes distributed document storage and editing engine, with a root node page and script mirroring content from as many locations as possible.
The Shelter team focused not on software, but on building physical resources with cheap, available materials to provide insulating protection against the elements. Using cardboard boxes and a vessel containing a human body’s worth of hot water, the team was able to simulate body temperature and heat loss by taking temperature readings at intervals. By adding a layer of tinfoil, they were able to significantly increase levels of insulation. On the ground, people will be able to take these commonly available materials to keep themselves warm. The Shelter plan depends on how many people need shelter, and the is cost based on that need. Plans for future implementation include how to waterproof materials. The team is looking into previous efforts on creating a cardboard canoe waterproof by soaking the cardboard in varnish.
The team is developing a tool for locating people in building collapse scenarios via device location. The premise of the project is that if the phone is working and findable, the person is also probably alive. Device Finder is working on triangulating location based on phone signal. The National Institute for Standards did tests in 2005 with radio transmitters and detected that building debris is a good blocker of RF. In order for the solution to work, the base station has to ping the device, and a WiFi network needs to be available. The device will connect to the strongest signal source, so if strong access points are nearby, phones will try to connect. The team plans to work on the project at Science Hack Day Dublin in the Spring.
Many thanks to TOG Dublin for being amazing hosts, to Tropo for supporting the event, and to Lisha for ordering us delicious food from 5 timezones away.