Intro Subway Map Applet How Credits Code/Docs Blog

Inspiration and Similar Stunts

Data Sources

The destination metrics (time, distance, fare) were scraped from the WMATA website using the incredibly useful POST-to-GET bookmarklet. (WMATA also has other nice maps on their site). This, like much of the other data, is just a collection of facts/common information and is therefore not copyrightable. The station geolocation information can be scraped from WMATA's website as well. Another way to find data like this: many Google maps widgets have their underlying data referenced transparently within the HTML. (Look for the embedded iframe and then search that file for .xml or load(..). Make sure that you may fairly use any data you acquire this way.

The background topo map is a public domain map (specifically, a DRG) made by the USGS -- but good luck getting anything from their site. Instead, I used the VA-MD-DC repository of USGS products from Radford University's Department of Geography and the Virginia Gazetteer at UVa, who ask me to cite them as "Virginia Economic Development Partnership, 2000 / U.S. Geological Survey, 19981031, USGS National Digital Orthophotography Program: US Geological Survey, Reston, VA." (It turns out that the best way to get GIS data is to hunt among the applicable state agencies, though the Internet archive has a reasonable sample.) The hardest non-programming part was patching the files together in Photoshop -- to get a seamless patch my base image is 18,000 by 18,000.

The background aerial photo was horked from Google Earth, but I'll replace it with a higher resolution (and public domain) DOQQ or landsat photo as soon as possible. Are you a GIS/Geography Expert? Contact me if you can help find a better (or otherwise interesting 3000x3000-or-higher-resolution map of the DC area.

I used a graphics program to pull dot locations from the clarified (London Underground-style) map at Wikimedia commons. This was the only data I had to type in to the computer. I wish I had found this map earlier -- it looks like it has better geographic fidelity which means less tearing.

You may download the image files (very large, register with each other) from Wikipedia:

There's more good stuff over at the Code / Docs page.