This is such a cool idea. I see more uses for it though. I live in a large city that is, unfortunately, fairly unfriendly toward commuter cyclists and cyclists in general. There are almost no bike lanes at all, and the ones that exist are small and dangerous . The drivers on main roads drive recklessly around cyclists, leaving as little room as possible. Despite the fact that it's illegal to cycle on the sidewalks, many drivers with no sense of how to share the road aggressively and maliciously cry out indignantly to demand that even the most careful and law-abiding cyclists "ride on the fucking sidewalk." Because of this, I often have to go to great lengths to plan my routes downtown and to popular areas carefully to avoid both busy roads and bad neighborhoods. Google maps never actually offers useful suggestions, but suggests routes that take me to the least bike-friendly roads. The software in this article sounds ideal. It would map out directions through less crowded and probably less car-centric areas. The potential here to make bike commuting in large, metropolitan, non-bike-friendly areas more pleasant is huge!
Yes indeed I do ask for a bicycle route. Sometimes it does alright, but other times it does terribly. For example: If there is a pseudo bike lane on a 40mph three-lane road (literally a regular car lane with a bike painted on it for good measure), and a quiet 25mph neighborhood route nearby that's a little bit out of the way, but goes the same way, Google maps will ignore the neighborhood route entirely in favor of the supposedly designated "bike lane" on the perilous and busier road.
It's an interesting idea, but I'm not sure they need to go to the trouble of crowd sourcing votes. You'd think that they could integrate data from Street View to determine, say, the number of trees, parks, or bodies of water along a route, and cross-ref that with Google Traffic data to filter by routes that have little traffic on them. Might be more difficult form a software perspective, but it would certainly be more expeditious when all is said and done.
Coding up a heuristic for things like "beauty" is hard. Training a classifier for beautiful images given enough images labeled beautiful or not would be relatively easy, but flickr tags are easier, and flickr tags worked out just as well for them in Boston as their crowdsourced voting in London.