4 ms·

A fair point, but I think we are way beyond Popular Science projections at this point.

by mason240 12y ago

marssaxman 12y ago
Helicopters actually exist. Self-driving cars are still nothing more than research prototypes.
- ghaff 12y ago
  I assume he meant in the sense of flying personal aircraft as helicopters were widely used in the 1950s. But I agree with your broader point. A lot of people aren't distinguishing the yawning distance between being able to operate on the highway with a backup driver ready to take over control and being able to operate just about anyplace, handle changes that aren't on a map, in any kind of weather, etc. even if there isn't a human inside the vehicle. I watched a recent video with John Leonard speaking at CMU and he had some interesting video showing some of the things he had to handle over the course of just a couple of weeks commuting. (Cops directing traffic around an accident, snow, left hand turn into heavy traffic, sun glare that made it hard to see a pedestrian or a light.) He makes the case that these are really hard problems that aren't anywhere close to being solved.
  - marssaxman 12y ago
    I was clearly too terse; the point I had in mind was that those Popular Science handwavings about the glories of the imminent personal-aviation future were actually more realistic than the current crop of handwavings about the glories of the imminent robot car future, because working commercially-available helicopters actually existed during the '50s-era personal-aviation hype. The engineering problems had been solved, so the leap of imagination only had to cross the safety issues and regulatory challenges - which, of course, ultimately proved intractable. With robot cars, in the year 2015, the imagination must jump even further, because there is no robot car industry yet; all we have are prototypes. Maybe those prototypes will become real products and maybe they won't; maybe those products will make it through all the non-technical hurdles necessary to make them a success, and maybe they won't. In any case, the people expecting robot cars to have a significant effect on our transportation infrastructure in a mere ten or fifteen years are being wildly optimistic, and the people who claim that we can stop investing in trains and buses because ubiquitous googlemobiles are just around the corner are... well, I'll be polite and stop talking about them.
    - ghaff 12y ago
      I fully agree. We're probably pretty close to having a lot of essentially convenience autopilot functions--however we decide to allow those to be phased in. (And however they're actually used.) But the "significant effects" require something more along the lines of robo-Uber and that's multiple decades out. It's worth noting that we already have "self-driving" cars. They just have organic computers as drivers. And you see quite a few of them--especially in Asian cities where there's a large wage gap.
- shard972 12y ago
  Do u even news, elon musk is pushing through an update in Q3 that will upgrade my car to a self-driving car.

A fair point, but I think we are way beyond Popular Science projections at this point.

marssaxman 12y ago

Helicopters actually exist. Self-driving cars are still nothing more than research prototypes.

ghaff 12y ago
I assume he meant in the sense of flying personal aircraft as helicopters were widely used in the 1950s. But I agree with your broader point. A lot of people aren't distinguishing the yawning distance between being able to operate on the highway with a backup driver ready to take over control and being able to operate just about anyplace, handle changes that aren't on a map, in any kind of weather, etc. even if there isn't a human inside the vehicle. I watched a recent video with John Leonard speaking at CMU and he had some interesting video showing some of the things he had to handle over the course of just a couple of weeks commuting. (Cops directing traffic around an accident, snow, left hand turn into heavy traffic, sun glare that made it hard to see a pedestrian or a light.) He makes the case that these are really hard problems that aren't anywhere close to being solved.
- marssaxman 12y ago
  I was clearly too terse; the point I had in mind was that those Popular Science handwavings about the glories of the imminent personal-aviation future were actually more realistic than the current crop of handwavings about the glories of the imminent robot car future, because working commercially-available helicopters actually existed during the '50s-era personal-aviation hype. The engineering problems had been solved, so the leap of imagination only had to cross the safety issues and regulatory challenges - which, of course, ultimately proved intractable. With robot cars, in the year 2015, the imagination must jump even further, because there is no robot car industry yet; all we have are prototypes. Maybe those prototypes will become real products and maybe they won't; maybe those products will make it through all the non-technical hurdles necessary to make them a success, and maybe they won't. In any case, the people expecting robot cars to have a significant effect on our transportation infrastructure in a mere ten or fifteen years are being wildly optimistic, and the people who claim that we can stop investing in trains and buses because ubiquitous googlemobiles are just around the corner are... well, I'll be polite and stop talking about them.
  - ghaff 12y ago
    I fully agree. We're probably pretty close to having a lot of essentially convenience autopilot functions--however we decide to allow those to be phased in. (And however they're actually used.) But the "significant effects" require something more along the lines of robo-Uber and that's multiple decades out. It's worth noting that we already have "self-driving" cars. They just have organic computers as drivers. And you see quite a few of them--especially in Asian cities where there's a large wage gap.
shard972 12y ago
Do u even news, elon musk is pushing through an update in Q3 that will upgrade my car to a self-driving car.