What is a basemap & how to choose the best one for your data
One of the most popular ways to make data accessible to people is to create a map. It is an interactive and intuitive way to integrate data into familiar concepts, whether you want to show locations, highlight evolutions using heatmaps, or display all the cities that voted for a left-wing candidate this year.
However, data on a map are just things drawn in the 2-dimensional air, and they need a canvas to be drawn on (roads, seas, city boundaries…). Why? Because, unfortunately, humans are still very bad at knowing that latitude 48.858333°, longitude 2.294167° is actually the Eiffel Tower (you know, between the patch of grass at 48.832,2.315353 and the 48.86123,2.282235 river).
As such, the canvas itself (or “basemap“) is mandatory, and has the heavy responsibility of being “under” your data and providing the geographical context; but many people tend to just pop the first basemap that comes around, and forget about it. It is actually easy to spend a little more time to find the right basemaps for your data, and make the data shine even more. OpenDataSoft makes it easy for you to integrate specific basemaps for your data.
Giving context to your markers with basemaps
Currently, apart from a few big players like Google, most of the basemaps you’ll find on the web are built using data from OpenStreetMap, a free project dedicated to gathering enough data to build a highly-detailed map of the entire world. Chances are that your area is already fully available inside OpenStreetMap (browse and see for yourself). This means that depending on the basemap, you could display very detailed information like trees, restaurants… For example, the OpenStreetMap project provides a few very specialized basemaps for transports or cycle roads, and you could use them to highlight very specific datasets:
OpenStreetMap Cycle Map
OpenStreetMap Transport Map
But you have many choices if you just want to show what people would consider as a “normal map” (which often means “like Google Maps” or “like Apple Plans”), and you’ll probably recognize the following from many maps on the web:
You can also find “special” basemaps, such as the “Pirate” map and the Comics map from Mapbox. To be honest, it is quite hard to use in your everyday life, but you may feel the need to find a reason.
Finally, there is the kind of basemaps where you try to find your house from above: the aerial or satellite views. This is a bit rarer for a few reasons: first, it can be expensive when done properly (did you ever wonder why no satellite view had clouds?) so it’s harder to offer for free; and second, it is very high on contrast and has a lot of different colors, so it is hard to display additional data on it and keep it visible. But still, providers like Mapbox have such available basemaps, sometimes with overlays for roads and boundaries.
Mapbox Streets Satellite
Neutral basemaps for analytical data
Maps are not only useful to locate points, they are also very powerful support for analytic representations. You can represent your city’s districts and color them by population density, display a heatmap of your country’s births after the Superbowl. Or you can show your city’s traffic using red and green roads. In these cases, you’re playing with different colors, or a single color gradients; for the sake of clarity, you don’t want to display this on a map with differents colors that may interfere with your analytics’s colors. Providers like Mapbox offer “neutral” maps that are perfect supports for these kinds of visualizations.
A lot of maps can benefit from such basemaps, as it would make them tremendously more clear and explicit.
A map canvas just for you
Sometimes, you want to have a basemap that is yours. Maybe you want a regular map but with your own colors; maybe you want a map that shows certain things and hide others to make it a better support for your data (for example a transport company may want a railroad-centric map to highlight their operational internal data). As basemaps are a well-established standard, there are a few different ways to create your own map, although all of them are a bit technical and require some non-trivial operations.
The easiest way (as far as we know) is Mapbox Studio which takes care of many things like hosting, but still requires you to learn a bit of CartoCSS (“CSS for maps”). There are free alternatives (free as in beer and free as in speech) to build your own stack, like Kosmtik + Tilestache. The point is, these basemaps can be seen as styles applied on top of cartographic data (e.g. from OpenStreetMap), and can be done by anyone with a few technical chops and enough time.
There are also a few other very specific tools to do things such as transforming an image into a map, which is what we’ve done for our French Cheese map using a software named MapTiler; you just have to send an image, and help the software by indicating which part of the image matches which real-life location.
Basemaps in OpenDataSoft
If you are an OpenDataSoft customer, it is easy to configure one or more basemaps for your data. In your Domain page, you’ll find a place to add and edit your basemaps: by default, you can choose between a few free basemaps; you can create a Mapbox account to benefit from the Mapbox integration and add their 14 basemaps to that choice. You can also configure your own basemap if you have the URL, or even a WMS service; just don’t forget to add attribution if the basemap requires it.
Want to leverage your data with APIs?
Grab our free 10-step guide now! It is loaded with hands-on advice on how to properly start your API project.