I knew that I wanted to play around with a mapping application for the last project in my writing seminar, so I started with TileMill. I was then led to the vortex that is MapBox Studio, and have compiled some of the lessons I’ve learned through this app:
- Finding Data
I had originally wanted to make a map with information on Isla Vista, the small college town I live in. Finding large sets of data points for IV was rather unfruitful, so I moved on to a bigger city.
Increasing numbers of cities and counties are providing open data resources for developers, analysts, and residents. Socrata is a pretty cool platform that has partnerships with San Francisco, New York City, and Seattle, just to name a few cities.
From SF Open Data, I found a CSV of all the Mobile Food Permits, in order to map out food trucks/carts in the city and different neighborhoods. I later realized that many of these points were not the actual locations of the food trucks/carts, but the residential homes of the people who applied for the permits. So I had to double-check the business addresses, and reconfigure the location coordinates. Luckily, there’s a resource for that.
In addition to the address flub, there was the issue of revolving schedules – many of these trucks would be at food truck events like Off the Grid and SoMa Street Food on certain weekends, or at different locations on different days.
In conclusion – pick your data set wisely!
2. TileMill to MapBox Classic to MapBox Studio.
I found out that TileMill was no longer in active development/was replaced by MapBox Studio. If you’ve taken any introductory GIS courses, the MapBox interface feels very similar with layer options. You can upload your own data sets (in this case, I uploaded my Mobile Permits CSV) to your account in order to reference it at any time.
I ultimately eschewed both TileMill and MapBox Studio to work with MapBox Editor – it was the most user-friendly and straightforward out of the options. You’re able to toggle between data, saved projects, and styles, while adding pop-up information to data points. I wanted to include each truck/cart’s address, schedule, and featured menu items.
I would recommend using MapBox Studio for more static data sets – the MapBox site has some example projects in their gallery, which includes a visualization of trees in New York. You can add a small legend in MapBox Studio, but if you wanted to include more information for each data point, it can get cluttered easily.
3. Limits of Open Data Set
While the Mobile Permits CSV file was a good starting point, the varied levels of online presence for each business played a big role in how much information I was able to provide. What I appreciate about this data set, in comparison to Yelp reviews, was that it included vendors from the high-profile trucks featured at SoMa Street Food and Off the Grid, to the family businesses with little to no web marketing.
Overall, I found MapBox to be a pretty user-friendly platform for my beginning adventures in mapping and data visualization. Excited to tinker around with it some more.