Monday, August 07, 2023

30 Second Data Viz with OSM GPT

I reviewed OSM GPT for the first time earlier today and I have to say I was very impressed with how easy it is to use in order to extract data from OpenStreetMap. 

OSM GPT's natural language interface for searching OSM is great for geographical searches - for example to find all cafes within 1,000 meters of a location ('get cafes within 1000 meters'). OSM GPT can also be used for creating some very quick data visualizations - by using different colors to display the results of two or more different searches.

The map in the screenshot above shows all Streets (colored green) and all Avenues (colored yellow) in Miami. The map illustrates perfectly how all roads named 'Street' in Miami run almost exclusively west to east while all 'Avenues' run north to south. It took me less than 30 seconds to create this visualization in OSM GPT by typing in two queries:
"get roads named 'Street'"
"get roads named 'Avenue'"
I then used the eyedropper tool in OSM GPT to color Streets green and Avenues yellow.
Here is similar map showing the orientation of streets and avenues in Manhattan. This time all the 'Streets' are colored blue and the 'Avenues' are yellow.

If you have trouble getting OSM GPT to run these queries in your town you can click the 'Manual Query' button instead and copy and paste this query to search for all roads named 'Street':

out skel qt;

Just change the word 'Street' to 'Avenue' to view all the roads named 'Avenue' (or to 'Lane', 'Alley, 'Close', 'Hill' etc for other alternatives).

In the last couple of years I've seen quite a few interactive maps which attempt to show how much inner city geographies are given over to cars by highlighting parking lots on a map. The screenshot above shows an area of Los Angeles after running the query 'get all parking'. This data viz took me only a few seconds. You can view how much of your town is a parking lot by simply centering OSM GPT on your address and typing in 'get parking'.
This map shows the space dedicated to parking lots (red) and to parks (green) in an area of Los Angeles.

No comments: