More saturated (darker) colors indicate a higher probability of tweets containing the text of the selected beer type. You can visually explore a variety of beers by selecting a new beer. By default, a given beer is normalized by the entire tweet population. You can also compare two different beers. Hexagons without a significant number of observations/tweets do not show up. That's why some beers have more coverage than others.
Utilizing the Leaflet library, the map aggregates point-level data into .35 degree hexbin polygon units to show the spatial distribution of the tweets related to specific beers. A single beer normalized by the tweet population is encoded using a 5-class sequential color scheme, while the comparison of two beers uses a 5-class divergent color scheme. The map is also projected into an Albers equal-area projection (with the help of CartoDB's cardodb.proj extension).
Data for this map was pulled from the Dolly project. We then wrote scripts in R and Node.js to process and aggregate totals into hexbin polygons, and encoded the results as GeoJSON, which were then loaded into the map client-side at runtime.
The Twitter API
A write-up and some static images are available at Floating Sheep.
The code for this project is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unsupported License and hosted as a New Maps Plus GitHub repository.
We apologize, but this thematic web map is best experienced on a tablet or PC.
While this is a terrible practice in most web design cases, we also believe that some interactive thematic maps don't really belong on smaller phones and handheld devices.