I recently came across a reference that stated over 100 streets in Paris are named after mathematicians. Having never heard that before, I decided to investigate and see what that looked like.

The following site had tallied 91 streets that were named after mathematicians and I thought I’d see what they might look on a map of Paris itself.

http://www-groups.dcs.st-and.ac.uk/~history/Honours/ParisNames.html

.The next step was to hopefully find some source of (hopefully) open data that represented Paris streets. I was able to find exactly this at the following site (after some Google translation; I don’t speak French beyond some intermediate menu ordering skills)

https://opendata.paris.fr/explore/dataset/voie/export/

I grabbed a shapefile from here and imported the street data into PostgreSQL using the shp2pgsql utility to do so (see here for specifics on using the utility). Now I had a table containing GIS information for all Paris streets, from which I needed to be able to filter those streets which were named after mathematicians.

I had previously imported the list of the 91 streets named after mathematicians into its own PostgreSQL table, so now naively thought a quick inner join of the street data to this table would get the GIS data for the 91 streets. But , of course, this is not the way it works when joining 2 disparate sources of data. It’s not the way it EVER works. Surprisingly, I was able to retrieve 80 of 91 just with a join on name, but this still meant I had to reconsider just attempting a join as my source. As the data was small enough, I just chose to add a column to the overall street table indicating whether it was named after a mathematician. Then, I could automatically update the 80 streets indicated above and , with a little manual intervention, update the remaining 11. If there were thousands of such streets, this would require some sort of more sophisticated attempt at pattern matching. As it is, a few SQL statements and I was able to have 86 of the 91 streets readily identifiable.

The following from the original list of 91 streets will not show on the map, 4 because I couldn’t match to an actual street and 1 (Emmy Noether) because it appears to be out of city limits. Still, we will be mapping 86 streets named after mathematicians.

Rue Arago

Emmy Noether – Out of city limits

Square du General Morin

Square Oronce Fine

Square Charle Hermite

Without further ado, here is the map of Paris streets, showing all streets in light blue and those named after mathematicians in dark red. I used QGIS with PostgreSQL data as a back end. Can we make this look nicer ? I bet we can !

Click for a larger view.