What you need to know
Can open data change the way citizens respond to disasters?
In the wake of 9/11, the U.S. government began urging citizens, “if you see something, say something,” but whatever got reported was shunted into the black box of national security, never to be seen again.
One startup called GeoThings is trying to bring about a new paradigm of disaster reporting in which anybody can report a problem and teams of volunteers can map it out for anybody to use or solve. GeoThings’ co-founder and CEO, the ironically named Slayer Chuang (莊國煜), told The News Lens that he saw the potential to work on this problem several years ago when he heard about uncoordinated responses to disasters.
Taiwan, Chuang said, experiences all sorts of “complicated disasters” like typhoons and earthquakes, and the country is relatively well prepared.
The trouble in Taiwan – and the world, really – is disparate methods of communication; different local governments and communities have their own LINE and Facebook groups and several kinds of social media that is difficult to coordinate. The government can send out text messages to the whole country, people mark themselves safe on Facebook, local Facebook groups can feature photos of danger spots during a typhoon – but this information is disjointed, proprietary, and much of it goes unsaved. Most disaster response is then a top-down effort led by the government or a few disaster response firms, leaving citizens reliant on a central reporting agency.
GeoThings’ main product is GeoBingAn, an app that allows users to report problems from their phones for automatic collection in a database that will mark their location and any details they add. Their database is then freely available to download, and anybody can delve into their growing list of reports, from flooded roads to injured kittens.
There are a few users who clearly don’t understand the point; some people included photos of themselves picking oranges. Playing around with the app, I accidentally told the world I needed to use a toilet (one of the default options) and found the message very unintuitive to retract.
Outliers aside, most of the data could be useful to researchers, government officials or concerned citizens.
If GeoThings can pass the great filter of big data and convince at least a critical mass of people to consistently use the app, they would have a data set that can be freely downloaded and used by the public and governments alike to map and respond to disasters. The Taoyuan City government has urged citizens to use the app; 1999 (Taipei’s citizen hotline) and the Taipei City police radio station are frequent data contributors.
GeoThings are not the first to allow people to become amateur cartographers, but their product places a strong emphasis on recording disaster-prone areas and keeping their data public. Their technologies rely on map database OpenStreetMap and Mapbox, largely open-source alternatives to the global giants Google Maps or Apple Maps.
But who is buying? Mostly humanitarian and development organizations. Their largest agreement is with Philippines-based Asian Development Bank, and they have also been involved in disaster risk reduction (DRR) in areas such as Bangladesh, the Philippines, and Fiji.
Uses vary depending on the geography. Chuang said that in places like the U.S., disaster response typically revolves around evacuation, but places like Fiji or Taiwan have less usable space to evacuate to. GeoThings is working on using an SMS standard that would allow a user’s location to be transmitted along with a text message. Mobile data is often scarce after a disaster, and sending a location by traditional means can sap precious time and battery power. Chuang says this technology has uses far beyond disaster relief, “Say you’re in a country and don’t speak the language. You would be able to send a text message with your location to a taxi driver or guide.”
Chuang added: “One problem in many developing countries is that there is no standardized address system; packages are given to a local post man who knows the community, who can then use context to get them to their destination.”
Normal databases are therefore useless to rescuers; if you don’t know the area, then an informal address will be unhelpful in locating an actual location, and most people are unable to navigate using a coordinate system.
In mapping, language is a problem that takes some work to overcome; many mapping tools don’t play well with Bengali or even Chinese. GeoThings gets around this by relying on crowdsourced mapping based on data collected from their app. They host occasional “mapping parties” in which the details of specific areas are ironed out from raw data and usable maps are drawn.
If they can iron out the bugs in the system, GeoThings could become a source of information for decades to come, not just in Taiwan, but in other disaster-prone areas around the world.
Editor: David Green