Rebecca Rosen for The Atlantic on why maps aren’t the best interface all of the time:

Think of it this way. In the days before online trip planners and GPS, if you wanted to know how to get from point A to point B, you would look at a map and trace out a route. But these days few people would use a map that way (I still do just because I enjoy the process but I think I’m in the minority). Instead, they would plug in their request and an algorithm would spit out a route for them. The route would appear on the map, but the map is no longer the tool for finding that answer.

In other words, just because the data has latitude and longitude attached to it, which seems like everything these days, you don’t need to automatically assume that you should throw it on a map.


  • Martin von Wyss July 12, 2012 at 3:50 am

    True, to a degree. But if you want to understand spatial data, or at least step back and admire it, then the best way to do that is throw it on a map. And that’s not that hard, thanks to the many tools out there now. Data are more fun when their spatial component is visible.

  • Printed route directions aren’t much help when you’re diverted or take a wrong turn.

  • Good points above. Maps to me have always been a way to interact with data. They are a User Interface through which we interact with our location, destination and POIs along our route. I don’t see maps losing relevance just because algorithms are more efficient at searching the map than we are. Underneath that GPS device is still a map.

    • Absolutely love this line: “Maps to me have always been a way to interact with data. They are a User Interface through which we interact with our location, destination and POIs along our rout”

  • Isn’t this just a different way to use or a different tool to apply to a map?

  • I am not disagreeing but I don’t think the example of not using paper maps to determine the best route is very persuasive. We have lots of algorithms that are superior to eyeballing data on a map but that isn’t what *presenting* the data in map form is about — you can do Moran’s I without ever seeing the data mapped (you do lose a lot of information that way but you can still determine if there is a statistically significant spatial pattern).

    You get a different appreciate for the data and analysis when you can see it visually. The question about mapping data really should be if the question is spatial. If you aren’t answering a question that is spatial, presenting the data, in map form adds nothing to the analysis so don’t do it, which is a kind of question that should be applied to every sort of data visualization (does this help illustrate the answer the question posed?).

    Back to the route-finding example, I depend on Google Maps to find the best route for me but I always look at the map of the route not just the list of turns because visually seeing the relation between where I am and where I am going gives me a better understanding of the directions that just reading a list of left and right turns (in fact, since I have trouble connecting the words left and right with the actual relationship of left and right I am likely to get lost depending on the text description rather than the visual depiction).

  • I once went on a kayaking expedition to Papua New Guinea. Had a map (though great swathes were blank with the words “obscured by clouds”). Anyway, we ended up walking out of a river gorge and armed with our map tried to ask the natives where the nearest road was. They had never seen a map. Couldn’t read a map and had absolutely no idea what it was trying to say. They just pointed vaguely over a mountain and said “Road not far. Two days”. It was a great example where people sometimes have completely different perspectives.

    ps next time I wasn’t so stupid and we bought a GPS. That way we could use our maps even if the natives couldn’t. Did you know you need to be able to see a certain number of satellites to fix a position? And you can’t see any sky in the rainforest. So it was back to a man with tattoos and machete: “How far is the road?” and waiting for the dreaded “Not far”.

  • Just to add another data point…I use the GPS on my Android smartphone frequently (Tokyo has a strange street system–even taxi drivers rely on GPS). While I do use the Google Maps algorithm to determine the quickest or shortest route, I almost never look at the list of directions given by Google Maps. I always use the route overlaid on the map.