Plotting information — say survey data in Pakistan’s Federally Administered Tribal Areas or election results in Afghanistan — on any kind of map adds critical geo-context to the data. These maps quickly become move powerful when you start adding more custom overlays, showing data like where different ethnic groups live, high incidents of corruption, or more complex visuals like the number of deaths per drone strike in Pakistan and which U.S. president ordered it.
What is really amazing is how accessible it is now for people to make custom maps to be able to tell more complex stories with data. Specifically, tools like Google Maps, OpenLayers, and Polymaps have made basic web mapping ubiquitous by making it simple to drop a map into a website, and their APIs open the door for everyone to customize maps by adding custom layers.
The trick now is to radically reduce the barrier to entry for making these overlays and custom base maps. This is the exact goal we had for our recently released TileMill, which is fully open sourced and built on top of fantastic open source tools.
TileMill is a modern map design studio that lets you design maps for the web using your own data or any publicly available data set. What makes TileMill unique is that it allows anyone who understands the idea behind CSS in web design to quickly and easily design custom maps.
Here is a look at using TileMill to make a custom baselayer map of Local Government Areas in Nigeria. This map helps visualize survey data around the upcoming presidential elections on April 9th.
Custom maps created in TileMill can be used on the web and also exported to run on mobile devices as an MBTiles file, an open format that packages up tiles to make them more portable and pluggable.
You can see TileMill in action in this video, where it’s used to create a map of Amtrak routes in the United States, and then exported for use on an iPad:
You can download TileMill and read documentation on how to use it at TileMill.com. If you run into any bugs or have other feedback, please let us know in the issue queue.
And finally, a big thank you to the Knight Foundation for funding much of this work.
