4 ms·

I hope for the day where all cars are self-driving. People will cycle more and 40000 families a year will be saved in the US alone.

akira2501 3y ago
Motorcycle fatalities are 15% of the total and pedestrians are another 16%. Most fatalities are single vehicle accidents. That 40000 figure is not entirely due to bad drivers, and expecting a single change to solve all problems is very short sighted.
- throwaway5959 3y ago
  Most of them are though. People can’t get off their phones, they’re addicted. Every time I see a car swerving in front of me and I pass, they’re on their phone. We have to figure out self driving cars because humans can’t disconnect and focus on their multi-ton vehicle.
  - seanmcdirmid 3y ago
    [flagged]
    - konschubert 3y ago
      Those deaths I heard about in the past year were all due to careless car drivers. Maybe it's different in the US.
      - seanmcdirmid 3y ago
        You might want to look into the, more deeply. Anti-car advocates (here ant least) are lumping in “high” pedestrians deaths with the cases you are expecting, to inflate their case of an increase due to traditional distracted driving. I don’t think it’s a coincidence that the increase in pedestrian deaths corresponds to the same timeline as the current fentanyl epidemic. But you have to go case by case to see where your local increase is coming from.
        konschubert 3y ago
        I see your point but I am not sure where you are going with this? Even if only 50% of the pedestrians killed were sober - isn't that enough of a crisis to act? 1500 children younger than 14 years died in traffic in the US in 2021. https://injuryfacts.nsc.org/motor-vehicle/overview/age-group-comparisons/#:~:text=In%202021%2C%20the%20number%20of,vary%20by%20type%20of%20crash https://injuryfacts.nsc.org/motor-vehicle/overview/age-group....
        seanmcdirmid 3y ago
        It’s the difference between saying “the crisis is getting worse!” And “we have a new crisis to deal with!” Also, understanding the components of the problem lead to you know, figuring out how to fix it, eg: https://www.kuow.org/stories/pedestrian-deaths-climb-in-seattle-despite-city-s-pledge-to-eliminate-them https://www.kuow.org/stories/pedestrian-deaths-climb-in-seat... I said this many times in other posts lately, but if Americans wanted to get serious about this problem, they would need a multi-faceted approach like the Dutch rather than just hoping drivers do the right thing (without focusing on road redesigns or pedestrian behavior also).
  - nicbou 3y ago
    You see even more as a motorcyclist because you have a direct view into vehicles. In Germany it's honestly not that bad, but a lot of people text as red lights.
- konschubert 3y ago
  Pedestrian fatalities - they happen when drivers speed, run red lights, hit pedestrians in crossings while turning. Self-driving cars will probably eliminate 90% of these. Same for people who are killed on bikes. With Motorcycles, I don't know. But I bet that at least half of these would also be eliminated if car drivers followed the rules.
  - t-3 3y ago
    Statistics I've seen have the pedestrian being inebriated in 40-50% of incidents. I know some scatter-brained people who have been hit multiple times while crossing intersections/parking lots/driveways - they just don't pay attention or look both ways. Self driving won't necessarily prevent those.
  - akira2501 3y ago
    They happen on class b highways at night. They often get hit from behind. There are more than just human factors. Self driving cars can't control pedestrians or guarantee they will follow traffic laws or not be inebriated. Also weather happens and snow and ice change everything. Self driving is not magic. Sensors have limitations, and can even just get dirty. Drivers won't ever follow rules. Automaton that allows our even requires driver intervention will probably not offer any improvement in fatalities.
    - konschubert 3y ago
      I think driver intervention will go down once you tell ppl they can just stay on their phones all the time.
    - creer 3y ago
      A recent rider video in a Waymo car showed pretty outstanding handling of unpredictable pedestrian behavior. One example of pedestrians partly hidden and then darting across the path of the car from between parked cars; another of pedestrian making to dart through and not going through with it; another of pedestrians hesitating (or easy to confuse with hesitating). The response to potentially dangerous pedestrian situations was noticeable and impressive. IMO much better than many human drivers would have. The Waymo car was hyper-alert and taking in a lot of clues - absolutely at the level of a hyper-alert human.
al_borland 3y ago
How will self driving cars lead to more cycling? Automating car traffic will lead to more car traffic. Today the number of cars on the road are limited to the number of drivers. If the drivers are removed, there can be even more cars, just driving around empty to go get someone. Even if demand for transit remains the same, driverless taxis have the potential to double the number of trips that need to be take. From going to a place and returning home (2 trips), to empty to pickup, go to a place, empty to pickup again, and returning home (4 trips). At scale, this will be a nightmare. To reduce car traffic we need to build alternatives to cars (protected bike lanes, rails, etc). We will never get people away from cars by building more/better cars.
- konschubert 3y ago
  I think we are talking about two distinct policy goals here: 1) humans, including children, should be able to move through the public realm safely, no matter if they walk or bike 2) we do not want all the space in cities to be taken up by cars Self-driving cars help with reaching goal number 1 partially, but it is not sufficient. Bike paths and other non-car infrastructure is also needed. In terms of goal number 2, self-driving cars might open up a political chance to ban on-street parking. You're right that they alone won't fix the problem, but I think they don't make it worse either. Traffic in cities is already limited by capacity, not by demand.
- aoeusnth1 3y ago
  The number one threat to pedestrians and cyclists is cars. If the cars were all safer, it would be more appealing to walk and bike.
  - ehnto 3y ago
    I think there is a misunderstanding of why cars are dangerous to cyclists and how to fix it. There are only two proven methods, slow car traffic right down til it is essentially co-mingling with pedestrians, or separate the two types of traffic completely. Automating cars changes neither of those. Maybe your lycra clad road cyclist doing 40kmph could co-exist with cars passing them at 60kmph+ but if you wouldn't let your kids cycle there then it's not actually safe. I would not trust a kid riding in a painted bike lane next to Tesla, even if it is less likely to hit them due to distractions there is still a whole host of things that can go wrong and then you face the fundamental issue of a speed and mass mismatch. Changing the roads and infrastructure is the only way to avoid that fundamental physics problem.

akira2501 3y ago

Motorcycle fatalities are 15% of the total and pedestrians are another 16%. Most fatalities are single vehicle accidents. That 40000 figure is not entirely due to bad drivers, and expecting a single change to solve all problems is very short sighted.

throwaway5959 3y ago
Most of them are though. People can’t get off their phones, they’re addicted. Every time I see a car swerving in front of me and I pass, they’re on their phone. We have to figure out self driving cars because humans can’t disconnect and focus on their multi-ton vehicle.
- seanmcdirmid 3y ago
  [flagged]
  - konschubert 3y ago
    Those deaths I heard about in the past year were all due to careless car drivers. Maybe it's different in the US.
    - seanmcdirmid 3y ago
      You might want to look into the, more deeply. Anti-car advocates (here ant least) are lumping in “high” pedestrians deaths with the cases you are expecting, to inflate their case of an increase due to traditional distracted driving. I don’t think it’s a coincidence that the increase in pedestrian deaths corresponds to the same timeline as the current fentanyl epidemic. But you have to go case by case to see where your local increase is coming from.
      - konschubert 3y ago
        I see your point but I am not sure where you are going with this? Even if only 50% of the pedestrians killed were sober - isn't that enough of a crisis to act? 1500 children younger than 14 years died in traffic in the US in 2021. https://injuryfacts.nsc.org/motor-vehicle/overview/age-group-comparisons/#:~:text=In%202021%2C%20the%20number%20of,vary%20by%20type%20of%20crash https://injuryfacts.nsc.org/motor-vehicle/overview/age-group....
        seanmcdirmid 3y ago
        It’s the difference between saying “the crisis is getting worse!” And “we have a new crisis to deal with!” Also, understanding the components of the problem lead to you know, figuring out how to fix it, eg: https://www.kuow.org/stories/pedestrian-deaths-climb-in-seattle-despite-city-s-pledge-to-eliminate-them https://www.kuow.org/stories/pedestrian-deaths-climb-in-seat... I said this many times in other posts lately, but if Americans wanted to get serious about this problem, they would need a multi-faceted approach like the Dutch rather than just hoping drivers do the right thing (without focusing on road redesigns or pedestrian behavior also).
- nicbou 3y ago
  You see even more as a motorcyclist because you have a direct view into vehicles. In Germany it's honestly not that bad, but a lot of people text as red lights.
konschubert 3y ago
Pedestrian fatalities - they happen when drivers speed, run red lights, hit pedestrians in crossings while turning. Self-driving cars will probably eliminate 90% of these. Same for people who are killed on bikes. With Motorcycles, I don't know. But I bet that at least half of these would also be eliminated if car drivers followed the rules.
- t-3 3y ago
  Statistics I've seen have the pedestrian being inebriated in 40-50% of incidents. I know some scatter-brained people who have been hit multiple times while crossing intersections/parking lots/driveways - they just don't pay attention or look both ways. Self driving won't necessarily prevent those.
- akira2501 3y ago
  They happen on class b highways at night. They often get hit from behind. There are more than just human factors. Self driving cars can't control pedestrians or guarantee they will follow traffic laws or not be inebriated. Also weather happens and snow and ice change everything. Self driving is not magic. Sensors have limitations, and can even just get dirty. Drivers won't ever follow rules. Automaton that allows our even requires driver intervention will probably not offer any improvement in fatalities.
  - konschubert 3y ago
    I think driver intervention will go down once you tell ppl they can just stay on their phones all the time.
  - creer 3y ago
    A recent rider video in a Waymo car showed pretty outstanding handling of unpredictable pedestrian behavior. One example of pedestrians partly hidden and then darting across the path of the car from between parked cars; another of pedestrian making to dart through and not going through with it; another of pedestrians hesitating (or easy to confuse with hesitating). The response to potentially dangerous pedestrian situations was noticeable and impressive. IMO much better than many human drivers would have. The Waymo car was hyper-alert and taking in a lot of clues - absolutely at the level of a hyper-alert human.

al_borland 3y ago

How will self driving cars lead to more cycling? Automating car traffic will lead to more car traffic. Today the number of cars on the road are limited to the number of drivers. If the drivers are removed, there can be even more cars, just driving around empty to go get someone. Even if demand for transit remains the same, driverless taxis have the potential to double the number of trips that need to be take. From going to a place and returning home (2 trips), to empty to pickup, go to a place, empty to pickup again, and returning home (4 trips). At scale, this will be a nightmare. To reduce car traffic we need to build alternatives to cars (protected bike lanes, rails, etc). We will never get people away from cars by building more/better cars.

konschubert 3y ago
I think we are talking about two distinct policy goals here: 1) humans, including children, should be able to move through the public realm safely, no matter if they walk or bike 2) we do not want all the space in cities to be taken up by cars Self-driving cars help with reaching goal number 1 partially, but it is not sufficient. Bike paths and other non-car infrastructure is also needed. In terms of goal number 2, self-driving cars might open up a political chance to ban on-street parking. You're right that they alone won't fix the problem, but I think they don't make it worse either. Traffic in cities is already limited by capacity, not by demand.
aoeusnth1 3y ago
The number one threat to pedestrians and cyclists is cars. If the cars were all safer, it would be more appealing to walk and bike.
- ehnto 3y ago
  I think there is a misunderstanding of why cars are dangerous to cyclists and how to fix it. There are only two proven methods, slow car traffic right down til it is essentially co-mingling with pedestrians, or separate the two types of traffic completely. Automating cars changes neither of those. Maybe your lycra clad road cyclist doing 40kmph could co-exist with cars passing them at 60kmph+ but if you wouldn't let your kids cycle there then it's not actually safe. I would not trust a kid riding in a painted bike lane next to Tesla, even if it is less likely to hit them due to distractions there is still a whole host of things that can go wrong and then you face the fundamental issue of a speed and mass mismatch. Changing the roads and infrastructure is the only way to avoid that fundamental physics problem.