Of all the datasets I want to carve up using D3, my Google Voice history is the most personal and interesting one I can think of. This project tells the story of who was important in my life at various points in time, as measured by how many SMS messages we exchanged in a given month.
When I run this visualization with my own GV account (all my texts have gone through GV since 2009), fascinating insights emerge. I see relationships blossom and fade. I see how important life events affect communication with others. I see how moving to a new city affected my friendships. I could stare at this thing for hours.
After cloning, put your Google Voice credentials in a JSON file called keys.json in the app/ directory:
{"email": "address@gmail.com",
"password": "123password321"}
Install dependencies (npm install -g bower if you haven't already installed bower):
npm install
bower install
Run node app/server.js and visit http://localhost:3000/index.html.
You may have to wait a minute or two if you have a lot of data. The data is cached as long as the server is running, so reloading the page should be quick. Color schemes are intentionally non-deterministic, so feel free to reload at will.