5 ms·

I'm willing to bet these people were following automated directions. My family and I were once driving from Chicago to my home town 6 hours to the southwest.

I'm willing to bet these people were following automated directions. My family and I were once driving from Chicago to my home town 6 hours to the southwest. A blizzard kicked up, and we were driving according to Google directions. It took us off of the main roads, and we found ourselves on unpopulated gravel roads in the middle of nowhere Illinois totally FUCKED of we had crashed. I've seen it over and over again. You trying to get somewhere. There are clear normal routes to that place. And then Google directions start to direct you to these fucking country roads to try and finesse some kind of "faster route" for you without even prompting you for a route change. It just directs you into some dumb fuck route that no one would ever take in the right mind. I've learned to watch Google directions like a hawk. I was once driving up from downstate, and it started to take us off the main road. I told my friend to stop and turn around, get back on the main highway. If we had taken the route Google recommended, we would have gone hours out of our way on gravel country roads to get around a bit of minor traffic on the main road.

carlivar 7y ago
Why did you obey Google and turn off the main roads?
adrianmonk 7y ago
Seems like Google (etc.) could add a "avoid roads that aren't well-traveled" feature with the data they already have. The data they use to figure out whether traffic is light or heavy (GPS location, speed, heading, path, etc.) could also be used to figure out whether roads are well-traveled. And they could even throw in month or season (or weather) as a variable. If the road is well-traveled in the summer but not in the winter, maybe there's a reason. This could be useful not only for winter but also for other issues like people who are driving an unreliable car or are low on gas (or not good about filling it up).
- nitrogen 7y ago
  During the last major fire season in California the traffic was incredibly slow on the main interstate I was on. Maps suggested a detour on a side highway because nobody was taking it. But it turned out that nobody was taking it because it was literally on fire, so I had to find my way back to the interstate.
  - adrianmonk 7y ago
    That one is going to be a little trickier to detect! I guess you could do some real time analysis and note that the number of people passing through this or that segment has dropped suspiciously below what it averages for that day of the week, time of day, etc. Which actually might be useful for other types of road closures like construction, so maybe not an unreasonable goal to aim for.
    - nitrogen 7y ago
      It seems like it would be worth including data from weather and emergency services. Another gripe I have is that it seems like pedestrian data is inadvertently used to determine road speeds in dense urban areas, so a block that is literally gridlocked for an hour will still be recommended by maps.
  - RandomBacon 7y ago
    Same thing in places that flood.

carlivar 7y ago

Why did you obey Google and turn off the main roads?

adrianmonk 7y ago

Seems like Google (etc.) could add a "avoid roads that aren't well-traveled" feature with the data they already have. The data they use to figure out whether traffic is light or heavy (GPS location, speed, heading, path, etc.) could also be used to figure out whether roads are well-traveled. And they could even throw in month or season (or weather) as a variable. If the road is well-traveled in the summer but not in the winter, maybe there's a reason. This could be useful not only for winter but also for other issues like people who are driving an unreliable car or are low on gas (or not good about filling it up).

nitrogen 7y ago
During the last major fire season in California the traffic was incredibly slow on the main interstate I was on. Maps suggested a detour on a side highway because nobody was taking it. But it turned out that nobody was taking it because it was literally on fire, so I had to find my way back to the interstate.
- adrianmonk 7y ago
  That one is going to be a little trickier to detect! I guess you could do some real time analysis and note that the number of people passing through this or that segment has dropped suspiciously below what it averages for that day of the week, time of day, etc. Which actually might be useful for other types of road closures like construction, so maybe not an unreasonable goal to aim for.
  - nitrogen 7y ago
    It seems like it would be worth including data from weather and emergency services. Another gripe I have is that it seems like pedestrian data is inadvertently used to determine road speeds in dense urban areas, so a block that is literally gridlocked for an hour will still be recommended by maps.
- RandomBacon 7y ago
  Same thing in places that flood.