We Need This: A Maps App That Algorithmically Finds You the Scenic Route


scenic-route-inline

liamgrantfoto/Getty



Here’s something nice: Researchers working out of a Yahoo lab in Barcelona are building a maps application that doesn’t just spit out the fastest path from A to B but instead endeavors to show you the most enjoyable one. What a concept! Is it too much to ask that the entire tech industry follow their lead?


The paper describing the effort, published last week, begins by stating what we already know: Today’s maps are very good at what they do. “When providing directions to a place, web and mobile mapping services are all able to suggest the shortest route,” our researchers write. This group had something different in mind–a map that can “automatically suggest routes that are not only short but also emotionally pleasant.”


The researchers–Daniele Quercia and Luca Maria Aiello from Yahoo!, and Rossano Schifanella from University of Torino, in Italy—started with crowdsourced data. Web users were shown two photographs of London streets and asked to choose which looked “more beautiful, quiet, and happy.” The researchers then aggregated those winning locations to create a new map of London–an alternate cartography weighted for human emotion. In tests, participants found that routes drawn from this map were indeed more pleasant, while adding just a few minutes to travel time.


Efficiency isn’t everything. We have the phrase “scenic route” for a reason.


The idea may be obvious, but it isn’t necessarily familiar. Thus far, the age of algorithms, data, and connectivity has been sold on the promise of efficiency. Silicon Valley is obsessed with doing more, faster, and our digital tools are built in its image. At one point, it was enough to be able put in your Chinese order online; now you can have a dude on a bike stop by pick it up for you. Next year, no doubt you’ll be able to wiggle your finger at your phone and get spring rolls air dropped to your apartment in minutes.


The cult of efficiency may not have any idol more awesome than Google Maps, which has made the world more knowable–and navigable–than ever before. It’s an incredible achievement by Google and its partners, and it’s an indispensable tool for the rest of us.


Still, efficiency isn’t everything. We have the phrase “scenic route” for a reason, and as Google’s driving directions increasingly become the only directions any of us ever think about checking, we risk losing sight of these alternate paths. That’s a shame. As this research shows, they’re often preferable. We should hope that the people building the next generation of apps and services spend more time considering these alternatives, whatever form they make take.


The Yahoo research offers a good template for how we might start to think about this challenge. For one, their project doesn’t upturn the underlying idea of routing–it just rejiggers it. The aim isn’t to turn every errand into a walking tour, it’s just to get you where you’re going a little more enjoyably. The only radical bit—which, of course, isn’t really radical at all—is the notion that we might be willing to trade a few minutes for a more pleasant experience overall. It’s still efficient, just with a dash of humanity.


It’s still efficient, just with a dash of humanity.


Smarter yet, the researchers are already exploring how you could do the whole thing at scale, algorithmically. The second phase of their experiment substituted Flickr data for the initial set of human-sorted images, the idea being that people are more likely to take pictures of interesting buildings and pretty streets than car-choked thoroughfares. Looking at 3.7 million photos of London and 1.3 million of Boston, they developed new maps computationally with promising results. In a Boston trial, a group of 50-some participants preferred the algorithmically generated routes to the shortest one every time.


Unfortunately, it doesn’t seem like today’s mapmakers are overly concerned with finding the “beautiful, quiet, and happy” routes through life. Instead, Apple and Google are focusing their energies on the the map wars’ latest frontier—indoors—with both companies scrambling to survey the places satellites haven’t yet been able to reach. To them, the dream is not just guiding you to the mall but directly to the store you’re looking for–and eventually, in retail establishments of a certain size, to the very item you desire. In other words: more exactitude, greater efficiency.


For their part, the researchers in Barcelona are currently building their own application. They want to refine their algorithms by getting them in peoples’ hands. The challenge there, of course, is getting people to try a new maps app in the first place. But who knows, as apps open up in new ways–say, with the arrival of “extensibility” in the next version of iOS–maybe we’ll someday be able to plug a third-party “scenic route” option right into Google Maps ourselves.


In any event, this sort of thinking is refreshing, and it could be fruitful in applications well beyond maps. Companies always talk about evoking emotions like “joy” and “delight” with their apps, but far too often they’re treated like byproducts, happy results of a well-polished UI. The far greater challenge is figuring out how we can incorporate these emotions into the basic functionality of our digital tools. It starts by acknowledging that the fastest route isn’t always the best one.



No comments:

Post a Comment