Visualizing Summer Travels Part 2: CartoDB

This post is part of a series on visualizing data from my summer travels.

I recently discussed OpenPaths and my goals in visualizing location data from my summer travels. In this post, I’ll explore visualizing this dataset with CartoDB. The OpenPaths data from my summer travels, which I’ll be working with in these examples, is available here and I discuss how I reverse-geocoded it here. CartoDB is a simple cloud-based tool for building web maps. You can import data through their web-based dashboard and quickly turn it into a dynamic map or visualization.

First, create an account and then go to your dashboard. Upload your location dataset as a new table, then click “map” to quickly generate a simple, default map of your dataset. The interactive map below displays my GPS location data over these two months as a set of static points:

This map is useful for displaying where I was, but from a high-level view it does not provide much other information. If you zoom in all the way to a place where I spent a lot of time (such as Barcelona, for example), you will find far more point data than you will in a place where I have only a single GPS point while in transit between locations. How can we better represent duration and paths through time and space?

Time and Flows in CartoDB

This next map uses CartoDB’s “torque” tool to create a dynamic visualization that does a better job of displaying paths and durations. In 30 seconds and 512 animation steps, it cycles through the approximately 1,800 location coordinates in my dataset. The dot pauses for a proportional amount of time in the locations where I spent more time, and zips quickly through the point data when I was in transit between locations.

These are two different ways of quickly and simply visualizing a location dataset. CartoDB is good for producing a nice-looking map in a short amount of time without requiring technical skills, but its customizability is in turn somewhat limited. Next, I’ll explore the Leaflet javascript library for mapping and visualizing this data, with a bit more for us to do under the hood.