4 ms·

Walkable cities are great and that's one way to reduce the impact of car accidents. Another way is to prevent the most catastrophic kinds of accidents from occu

Walkable cities are great and that's one way to reduce the impact of car accidents. Another way is to prevent the most catastrophic kinds of accidents from occurring in the first place by manufacturing the cars to be incapable of them. There's no reason, for example, that a car traveling on an urban street must dutifully obey the command of the buzzed teenager driving it to accelerate to 100 mph. We don't need to wait for full self-driving to make the cars more aware of their surroundings and incapable of doing things that are already illegal and absurdly dangerous. All new cars manufactured for use on public rights-of-way should immediately be throttled to the nation's highest speed limit on any road (85 mph in the U.S.). The most obvious kinds of awareness capabilities should shortly follow (e.g. am I on an Interstate Highway right now? No? Then I cannot travel at speeds that are only legal on Interstate Highways.) We don't manufacture any other product with buttons that invoke illegal functions and then spend millions on education and enforcement to persuade people never to press them. It is absurd to allow the manufacture of a product that is designed for breaking the law and that we know with great statistical certainty will used to do so.

vxNsr 3y ago
Have you considered the unintended consequences of this sort of authoritarian reactionary policy?
- dionidium 3y ago
  Manufacturing a car to travel only at the prevailing speed limit is "authoritarian" and "reactionary?"
  - malwrar 3y ago
    Sure, consider what would need to happen for a system like this to actually be implemented. How would the car know what kind of road it is on? How would the car know the appropriate speed? What happens to sports cars and other enthusiast vehicles? What happens with existing vehicles? All these questions raise privacy and reliability concerns in my mind, not to mention the distinctly weird feeling of some outside party deciding what I can and can not be trusted with. This feels authoritarian to me, and as a proposal feels reactionary in the sense that the solution is simply “ban it”.
    - xnx 3y ago
      Many cars already have speed governors built-in by the manufacturers. It seems reasonable that these limits should be set [at most] to the top speed allowed on US roads.
    - IshKebab 3y ago
      The car would not need to know why road it's on because he suggested limiting it to the national speed limit, not the speed limit of the road you are currently on. Existing vehicles would eventually be replaced. Sports enthusiasts would presumably get some kind of exemption. I can't imagine what kind of privacy or reliability concerns you are imagining from limiting cars to 85mph. However... while limiting cars to 85mph is very sensible, the EU has decided to flex their idiot nanny muscles and do something much more stupid. Continuing in the tradition of the headphone volume and cookie laws they have come up with an extremely annoying law that will achieve none of its goals. Basically they will use unreliable GPS to figure out the current speed limit, but then it won't actually limit your speed, it will just beep at you annoyingly. You can disable it, but it will re-activate every time you start the car. Can I get 1000 face palms for the EU? It's almost enough to make you support Brexit.
conductr 3y ago
Conceptually agree. Although my car and even phone GPS regularly think I’m on a road that I’m not actually on so I don’t think this would execute well. To give an example, when I’m on a highway bridge it things I’m on a city street below. Or when I’m on a highway it gets confused and thinks I’m on an access road. The difference is legal speed limits can be 40 mph in these cases and I’m not sure what would happen if car was randomly governing my speed based to a lower than expected speed and then bouncing back and forth. There’s probably a lot of situations where the tech we’ve become comfortable with still isn’t quite ready for this. I’m in the camp of believing speed limits aren’t inherently correct/safe except in trying to reduce variation of speed within the population of autos. I feel like this could increase variation.
- dionidium 3y ago
  > Conceptually agree. Although my car and even phone GPS regularly think I’m on a road that I’m not actually on so I don’t think this would execute well. This is a win in my book. If we can move past the conceptual/philosophical objections, then we can get into implementation details, which is a much better discussion to be having.
ryandrake 3y ago
Speed is an impact multiplier when the accident happens, but I doubt speed in and of itself is the root cause of most accidents. The article itself points to Speed causing 14% of costs vs. Alcohol + Distraction accounting for 49% of costs. Another study showed speeding was involved in 9% of property-damage-only crashes and 13% of crashes with injuries or fatalities in 2021[1]. I'd rather law enforcement spend their limited time cracking down on drunk driving and all the various distractions that actually tend to lead to accidents, rather than the easy job of "farming" people going 10mph over the limit for steady revenue, or nerfing cars to some arbitrary limit. And not just drunk driving. I took a bus ride the other day and just people-watched into cars, and about 60-80% of drivers were scrolling on their smartphones (in-hand) while driving. If you fined every single one of them $100 every time they picked up their phones, you could probably fully fund your city's public safety budget. Maybe the entire city's budget. 1: https://www.nhtsa.gov/crash-data-systems/crash-report-sampling-system https://www.nhtsa.gov/crash-data-systems/crash-report-sampli...
- dionidium 3y ago
  NHTSA says here [0] that it's a factor in 29% of fatalities. I think there's also reason to suspect that number is underreported. The responding officer has to make the judgement on the scene that speed was a factor and decide to write that down on the accident report. We should really be doing a lot more tracking of this kind of data using telemetry in the cars themselves (ideally as required by insurance). [0] https://www.nhtsa.gov/risky-driving/speeding https://www.nhtsa.gov/risky-driving/speeding
  - mcpackieh 3y ago
    A lot of speed-induced accidents probably happen below the posted speed limit, but above whatever speed is safe given the conditions of the road at that moment (wet, icy, foggy, etc.)
- mcpackieh 3y ago
  To borrow the form of the GP comment, there is no particular reason a phone should allow itself to be used when its owner is driving a car. Creating some arrangement of sensors and systems that could lock the phones of drivers may not be straight forward and would require buy-in from both phone manufacturers and car manufacturers, but seems possible in principle. The car could scan the drivers' face, then interrogate nearby phones, asking them if the driver's face matches the phone's owner. The phone could then lock itself if it does. There would be technical and privacy hurdles to overcome, but nothing impossible.
franga2000 3y ago
> We don't manufacture any other product with buttons that invoke illegal functions and then spend millions on education and enforcement to persuade people never to press them. There's one more. Just like cars, misuse of it is also the most prevalent in the USA, more and more are being sold, they're getting bigger and more dangerous, the usual excuse for it is freedom and the license required to operate one is (again, mainly in the USA) granted to basically anyone. It's like Americans are allergic to public safety...
- dionidium 3y ago
  This is an obvious comparison, but it fails because there is no conceivable way to disable the illegitimate uses of a firearm without impacting the functioning of the allowed uses. If we could invent a gun that would not fire in a bad shooting, but would still function in self-defense, then obviously we should do that! But of course we can’t. (It’s not even just a technological problem. It’s that the problem space is far too fuzzy to imagine doing this even in principle.) Speed, on the other hand, is discrete, measurable, and discernible. You can prevent a car from exceeding the speed limit without impacting any of its legitimate functionality. Crucially, you would only be disabling behavior that is already illegal.
  - johnnyanmac 3y ago
    You can argue the same part for cars. First of all, cars do have speed limiters built into them, but they are a matter of power, not objective speed. It'd be hard for a car on the hardware level to figure out the latter (and there are already many debates as is about software controlling your car's hardware. I'll skip over that for now). And secondly, that power is why it's hard to make this law: hills. What may be a limited max SPD of 80 on a flat plane can be below 60 on a steep hill, or over 100 going downhill. I believe that is why the compromise for a limiter is more around 120-130 MPH instead of something lower: it wants an average sedan to be able to not stall out going up a hill on a free way. Now granted, that hasn't been an issue for almost 2 decades. But it still impacts how they limit car speeds.

vxNsr 3y ago

Have you considered the unintended consequences of this sort of authoritarian reactionary policy?

dionidium 3y ago
Manufacturing a car to travel only at the prevailing speed limit is "authoritarian" and "reactionary?"
- malwrar 3y ago
  Sure, consider what would need to happen for a system like this to actually be implemented. How would the car know what kind of road it is on? How would the car know the appropriate speed? What happens to sports cars and other enthusiast vehicles? What happens with existing vehicles? All these questions raise privacy and reliability concerns in my mind, not to mention the distinctly weird feeling of some outside party deciding what I can and can not be trusted with. This feels authoritarian to me, and as a proposal feels reactionary in the sense that the solution is simply “ban it”.
  - xnx 3y ago
    Many cars already have speed governors built-in by the manufacturers. It seems reasonable that these limits should be set [at most] to the top speed allowed on US roads.
  - IshKebab 3y ago
    The car would not need to know why road it's on because he suggested limiting it to the national speed limit, not the speed limit of the road you are currently on. Existing vehicles would eventually be replaced. Sports enthusiasts would presumably get some kind of exemption. I can't imagine what kind of privacy or reliability concerns you are imagining from limiting cars to 85mph. However... while limiting cars to 85mph is very sensible, the EU has decided to flex their idiot nanny muscles and do something much more stupid. Continuing in the tradition of the headphone volume and cookie laws they have come up with an extremely annoying law that will achieve none of its goals. Basically they will use unreliable GPS to figure out the current speed limit, but then it won't actually limit your speed, it will just beep at you annoyingly. You can disable it, but it will re-activate every time you start the car. Can I get 1000 face palms for the EU? It's almost enough to make you support Brexit.

conductr 3y ago

Conceptually agree. Although my car and even phone GPS regularly think I’m on a road that I’m not actually on so I don’t think this would execute well. To give an example, when I’m on a highway bridge it things I’m on a city street below. Or when I’m on a highway it gets confused and thinks I’m on an access road. The difference is legal speed limits can be 40 mph in these cases and I’m not sure what would happen if car was randomly governing my speed based to a lower than expected speed and then bouncing back and forth. There’s probably a lot of situations where the tech we’ve become comfortable with still isn’t quite ready for this. I’m in the camp of believing speed limits aren’t inherently correct/safe except in trying to reduce variation of speed within the population of autos. I feel like this could increase variation.

dionidium 3y ago
> Conceptually agree. Although my car and even phone GPS regularly think I’m on a road that I’m not actually on so I don’t think this would execute well. This is a win in my book. If we can move past the conceptual/philosophical objections, then we can get into implementation details, which is a much better discussion to be having.

ryandrake 3y ago

Speed is an impact multiplier when the accident happens, but I doubt speed in and of itself is the root cause of most accidents. The article itself points to Speed causing 14% of costs vs. Alcohol + Distraction accounting for 49% of costs. Another study showed speeding was involved in 9% of property-damage-only crashes and 13% of crashes with injuries or fatalities in 2021[1]. I'd rather law enforcement spend their limited time cracking down on drunk driving and all the various distractions that actually tend to lead to accidents, rather than the easy job of "farming" people going 10mph over the limit for steady revenue, or nerfing cars to some arbitrary limit. And not just drunk driving. I took a bus ride the other day and just people-watched into cars, and about 60-80% of drivers were scrolling on their smartphones (in-hand) while driving. If you fined every single one of them $100 every time they picked up their phones, you could probably fully fund your city's public safety budget. Maybe the entire city's budget. 1: https://www.nhtsa.gov/crash-data-systems/crash-report-sampling-system https://www.nhtsa.gov/crash-data-systems/crash-report-sampli...

dionidium 3y ago
NHTSA says here [0] that it's a factor in 29% of fatalities. I think there's also reason to suspect that number is underreported. The responding officer has to make the judgement on the scene that speed was a factor and decide to write that down on the accident report. We should really be doing a lot more tracking of this kind of data using telemetry in the cars themselves (ideally as required by insurance). [0] https://www.nhtsa.gov/risky-driving/speeding https://www.nhtsa.gov/risky-driving/speeding
- mcpackieh 3y ago
  A lot of speed-induced accidents probably happen below the posted speed limit, but above whatever speed is safe given the conditions of the road at that moment (wet, icy, foggy, etc.)
mcpackieh 3y ago
To borrow the form of the GP comment, there is no particular reason a phone should allow itself to be used when its owner is driving a car. Creating some arrangement of sensors and systems that could lock the phones of drivers may not be straight forward and would require buy-in from both phone manufacturers and car manufacturers, but seems possible in principle. The car could scan the drivers' face, then interrogate nearby phones, asking them if the driver's face matches the phone's owner. The phone could then lock itself if it does. There would be technical and privacy hurdles to overcome, but nothing impossible.

franga2000 3y ago

> We don't manufacture any other product with buttons that invoke illegal functions and then spend millions on education and enforcement to persuade people never to press them. There's one more. Just like cars, misuse of it is also the most prevalent in the USA, more and more are being sold, they're getting bigger and more dangerous, the usual excuse for it is freedom and the license required to operate one is (again, mainly in the USA) granted to basically anyone. It's like Americans are allergic to public safety...

dionidium 3y ago
This is an obvious comparison, but it fails because there is no conceivable way to disable the illegitimate uses of a firearm without impacting the functioning of the allowed uses. If we could invent a gun that would not fire in a bad shooting, but would still function in self-defense, then obviously we should do that! But of course we can’t. (It’s not even just a technological problem. It’s that the problem space is far too fuzzy to imagine doing this even in principle.) Speed, on the other hand, is discrete, measurable, and discernible. You can prevent a car from exceeding the speed limit without impacting any of its legitimate functionality. Crucially, you would only be disabling behavior that is already illegal.
- johnnyanmac 3y ago
  You can argue the same part for cars. First of all, cars do have speed limiters built into them, but they are a matter of power, not objective speed. It'd be hard for a car on the hardware level to figure out the latter (and there are already many debates as is about software controlling your car's hardware. I'll skip over that for now). And secondly, that power is why it's hard to make this law: hills. What may be a limited max SPD of 80 on a flat plane can be below 60 on a steep hill, or over 100 going downhill. I believe that is why the compromise for a limiter is more around 120-130 MPH instead of something lower: it wants an average sedan to be able to not stall out going up a hill on a free way. Now granted, that hasn't been an issue for almost 2 decades. But it still impacts how they limit car speeds.