Reviewing the data visualisation articles that I’ve most liked over the last couple of months there seemed to be an American theme. Though one visualisation has made me wonder: even across the ocean in America do all roads lead to Rome?

All roads lead to Rome

Some data scientists at Moovel Labs took up the challenge of seeing whether all roads do in fact lead to Rome. Using open source tools including Graph Hopper, mapbox GL JS and mongoDB, they wrote an algorithm to see. After creating a beautiful map showing roads across Europe spidering towards Italy they turned their attention to other continents. They counted up 9 towns and cities in the USA called Rome and created the map view below:

Their algorithm took around 2 hours to determine which was the closest Rome and then calculate the shortest distances from all points in the US.

If you have time to explore their analysis it will be well worth your while … they also looked at how far you could get in given times from a selected point, creating what they called “Street DNA”. The below image shows the travel distances from Trafalgar Square in 15 minutes. It’s close to being the most symmetric of the cities considered.

I looked at the street DNA for my home in East London and got a far less symmetrical picture, although it was better when I switched from car to bike and then to walking.

Where’s the nearest grocery store?

Nick Feltron took up a similar challenge and looked across all of the US for the closest grocery store. This map nicely shows how sparsely populated the Western States are by comparison with the East.

American Immigration

Still on the theme of travel (tenuous link?), Alvin Chang of Vox, created an interactive slide show demonstrating how immigration to the US has changed over the past 200 years.  He includes a voiceover explaining how global and US politics influenced the changes.

What an average American does

Nathan Yau of Flowing Data took microdata from the American Time Use survey of 2014 which asked thousands of people what they did in a 24 hour period. He then simulated a day for 1000 Americans representative of the population, to the minute, using an incredible algorithm that tabulated transition probabilities between states.

See how nearly 20% of Americans are still sleeping at 8:45 in the morning! To me, an early bird usually up before 6am, this is quite astonishing.