5 ms·

3 years probation for a set of actions that resulted in a loss of life. I'm always aghast at this, we call things that happen on the road "accidents" as if it'

by buro9 3y ago

3 years probation for a set of actions that resulted in a loss of life. I'm always aghast at this, we call things that happen on the road "accidents" as if it's "oops, I killed you, sorry". The language of reporting isn't on the drivers "car killed pedestrian", rather than "driver killed pedestrian". The deaths by vehicles is astonishing, last year in the USA 42,795 people died in road traffic incidents ( https://www.nhtsa.gov/press-releases/traffic-crash-death-estimates-2022 https://www.nhtsa.gov/press-releases/traffic-crash-death-est... ), in the UK the number is 1,695 ( https://www.gov.uk/government/statistics/reported-road-casualties-great-britain-provisional-results-2022/reported-road-casualties-great-britain-provisional-results-2022 https://www.gov.uk/government/statistics/reported-road-casua... ). In the USA alone, that's 117 people per day. Virtually no-one who owned the set of decisions that led to those deaths face any consequence of any kind... for ending someone's life. When the article says "Getting behind the wheel of a car is a serious responsibility"... there are no real consequences for the individuals involved for their choices, and choices should have consequences. Driving too fast in rural areas, not paying attention, not keeping your car in good repair, being under the influence... these are all choices, and choices should have consequences. The consequences should be significant enough that they disincentivise that behaviour in everyone else, which means never being able to drive again, jail time, significant and meaningful financial penalties (with a large portion paid to the victims family). 3y probation, as an outlier exceptional example of consequences, shows the priority is on vehicle use rather than people's lives.

quietbritishjim 3y ago
> The deaths by vehicles is astonishing, last year in the USA 42,795 people died in road traffic incidents ( https://www.nhtsa.gov/press-releases/traffic-crash-death-est https://www.nhtsa.gov/press-releases/traffic-crash-death-est... ), in the UK the number is 1,695 ( https://www.gov.uk/government/statistics/reported-road-casua https://www.gov.uk/government/statistics/reported-road-casua... ). I'm not sure whether this was your point, but USA population is just under 5 times the UK population, but that US road deaths figure is more than 25 times the UK one. Of course, the nature of UK city centres and geographic size means that there is bound to be more travelling by car in the US in the first place. But at least part of the difference has to be a different emphasis on road safety in the UK compared to the US.
- lm28469 3y ago
  Distance driven per capita would be a better metric than population
  - eesmith 3y ago
    Including the worked-out answer in your comment would also be better. Or see https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r... which lists those numbers for 2019 and points out that population is only one of several relevant comparison: per 100k per 100k per 1 billion inhabitants motor vehicles vehicle km per year per year ---------- -------- ---------- UK 2.9 5.7 3.8 (in 2010) USA 12.9 16.1 8.3 (in 2022)
    - quietbritishjim 3y ago
      This is probably the best we can reasonably do but even this doesn't show the whole picture. I think it makes it look like most (but not all) of the safety difference is due to much less distance driven in the UK. But, given identical attitudes and driving style, I would expect accidents per km to be much higher in the UK because more time is spent on much slower urban roads in town centres (perhaps not big city centres so much) and UK suburban (which is much higher density => smaller distance than US suburban). So the spatial density of opportunity for vehicle/pedrestrian collision is much greater. Perhaps per unit time spent driving would be closer, but still not there. But that's all to say, I think the UK numbers are better than the USA ones by an even larger factor than they appear.
      - eesmith 3y ago
        > This is probably the best we can reasonably do but even this doesn't show the whole picture. Well, the numbers I gave only concern "deaths by vehicles", not "accidents". ("Accident" is a misnomer as it's a marketing term used to normalize car collisions - https://en.wikipedia.org/wiki/Accident#Accidents_by_vehicle https://en.wikipedia.org/wiki/Accident#Accidents_by_vehicle .) The death rate increases with speed, and I think US policies encourage higher speeds than the UK. UK cars also tend to be smaller and lighter, which reduces the size of the blind spots and the fatality rate of collisions. We can surely do better than my stats as there are a lot of publications on this topic.
knolan 3y ago
It seems that if you want to murder someone and get away with it you should do it in a car.
LatteLazy 3y ago
On the other hand, what's the point in locking up someone who (a) statistically will never do it again and (b) never intended or foresaw that this would happen?
- andrewfurey2003 3y ago
  I suppose people would take getting into and driving a car more seriously
  - LatteLazy 3y ago
    Would they? A lethal accident is something like once per million journeys. I don't think about that risk when I start my car. Adding a prison sentence would still not make me think about it, because it's such a tiny risk... Sorry if that's a bit blunt. It would definitely be better if people were more safety aware. Maybe making people watch videos like this as part of (re)training of drivers would be more effective?
    - badcppdev 3y ago
      If you knew that the penalty for killing someone while speeding was 10 years in prison then would you stay under the speed limit?
      - LatteLazy 3y ago
        No, No more than I currently do anyway. I already strongly don't want to kill anyone. But I also know the chances of me doing so are very very low. So that 10yrs is very heavily discounted by the tiny probability of it happening. I guess an analogy for this is that I don't change my behaviour after I buy a lottery ticket. Sure it is not technically impossible that I will be a millionaire soon. But I don't take the precaution of lining up a lawyer and a tax accountant because the chances of it coming to pass are so tiny...
NoZebra120vClip 3y ago
Imagine what the penalty would have been if the "backup driver" didn't already have a felony conviction record.
spoonjim 3y ago
Killing someone unlawfully should have a minimum of 10 Christmases in prison.
- Lolaccount 3y ago
  A 10 day sentence is pretty weak ...
friend_and_foe 3y ago
I think youre drawing the conclusions you want from the scenario. The truth is that Uber removed safety features from the car that could've stopped this and the car only recognized pedestrians if there was a crosswalk. If you ask me this is not the woman's fault, they're framing it like she was "at the wheel" because they want to shirk responsibility, and probably paying good PR money to these news rags to keep the narrative going too. Uber did this, they knowingly put an unsafe machine on the road and it killed someone. They put someone at a computer in a call center as a fall guy and it worked.
- buro9 3y ago
  I didn't draw conclusions about this specific instance. Consequences should exist, and should be a deterrence against unsafe behaviour no matter the party. Consequences for corporations should be more significant than for individuals IMHO. If Uber contributed to this in their training of their safety driver, the equipment in use, disabling built-in safety features, and so forth... then they should have penalties severe enough to adjust their behaviour too. My thinking is that the more vulnerable a road user, the greater the ownership is on the less vulnerable, meaning the greater the consequences of their choices. What I didn't do is limit "their" to just a driver / safety driver.
goodpoint 3y ago
"consequences" do not solve the problem because people cannot self-evaluate their driving skills and general attention level. In US getting a driving license is shockingly easy and cars are huge and inherently dangerous. You need train and trams.
unsupp0rted 3y ago
I don't see what making the woman do hard time for 10 years would accomplish. People kill each other with vehicles all the time, because apes that attempt to drive 2-ton metal canisters at 100 kph around other apes doing the same, and some apes on bicycles or walking by, will inevitably do something catastrophic. I don't really blame the apes at the wheel, since they're trapped in a system that is much larger than them. After a year or two of intentional driving experience leading to muscle memory, ape brains are wired to fall into comfortable inattention in these scenarios. What can you do, other than give up driving and give up walking where cars go? We weren't designed to do this safely.
nullc 3y ago
> In the USA alone, that's 117 people per day. Against how many passenger miles or hours?
acumenical 3y ago
Generally I agree with you, drivers get away with murder too easily, however in this case this woman's crime was not paying attention in a self driving car. What if instead she had been paying attention but applied the brakes too late, but it wasn't enough? What if she was paying attention but was confused by instructions from her employer to let the AI do its job? What if she had been in the same situation before but the car had previously stopped in exactly the same circumstances? Even engineers become confused by LLM and think that it's so quirky that ChatGPT "hallucinates" when that's the nature of the technology. So how would you expect a layperson to actually know when they should apply the brake pedal in a moving vehicle that's supposed to drive itself (maybe)? This is a bit more complicated than driving drunk and I'm fairly certain that the company responsible for the death did not seek a safety expert to copilot the vehicle, but rather any warm body willing to take an easy dead-end job.
- darkclouds 3y ago
  > This is a bit more complicated than driving drunk I'd argue no, manganese needs to be quantified in the blood stream when drunk driving imo, the state doesnt do that, but might if the doctor is called to take a blood sample having refused a breath test at the road side and/or in the police station. Manganese has a half life of 10seconds in the blood stream, and alcohol is very good at releasing it into the blood stream for elimination and thus deficiency. When is a drunk person or a baby or toddler, not suffering from manganism which can be likened to Parkinsons? Plus there are some gut bacteria which will produce ethanol from sugars, which is why some people always blow positive for drink driving even when they have not been drinking. Simply measuring alcohol is not enough, and if you happen to be able to blow you own trumpet, then many musicians will know that they can do a rebreathing trick, which means they will never fail a breath test, even if they fall out of the vehicle!

by buro9 3y ago

quietbritishjim 3y ago

> The deaths by vehicles is astonishing, last year in the USA 42,795 people died in road traffic incidents ( https://www.nhtsa.gov/press-releases/traffic-crash-death-est https://www.nhtsa.gov/press-releases/traffic-crash-death-est... ), in the UK the number is 1,695 ( https://www.gov.uk/government/statistics/reported-road-casua https://www.gov.uk/government/statistics/reported-road-casua... ). I'm not sure whether this was your point, but USA population is just under 5 times the UK population, but that US road deaths figure is more than 25 times the UK one. Of course, the nature of UK city centres and geographic size means that there is bound to be more travelling by car in the US in the first place. But at least part of the difference has to be a different emphasis on road safety in the UK compared to the US.

lm28469 3y ago
Distance driven per capita would be a better metric than population
- eesmith 3y ago
  Including the worked-out answer in your comment would also be better. Or see https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r... which lists those numbers for 2019 and points out that population is only one of several relevant comparison: per 100k per 100k per 1 billion inhabitants motor vehicles vehicle km per year per year ---------- -------- ---------- UK 2.9 5.7 3.8 (in 2010) USA 12.9 16.1 8.3 (in 2022)
  - quietbritishjim 3y ago
    This is probably the best we can reasonably do but even this doesn't show the whole picture. I think it makes it look like most (but not all) of the safety difference is due to much less distance driven in the UK. But, given identical attitudes and driving style, I would expect accidents per km to be much higher in the UK because more time is spent on much slower urban roads in town centres (perhaps not big city centres so much) and UK suburban (which is much higher density => smaller distance than US suburban). So the spatial density of opportunity for vehicle/pedrestrian collision is much greater. Perhaps per unit time spent driving would be closer, but still not there. But that's all to say, I think the UK numbers are better than the USA ones by an even larger factor than they appear.
    - eesmith 3y ago
      > This is probably the best we can reasonably do but even this doesn't show the whole picture. Well, the numbers I gave only concern "deaths by vehicles", not "accidents". ("Accident" is a misnomer as it's a marketing term used to normalize car collisions - https://en.wikipedia.org/wiki/Accident#Accidents_by_vehicle https://en.wikipedia.org/wiki/Accident#Accidents_by_vehicle .) The death rate increases with speed, and I think US policies encourage higher speeds than the UK. UK cars also tend to be smaller and lighter, which reduces the size of the blind spots and the fatality rate of collisions. We can surely do better than my stats as there are a lot of publications on this topic.

knolan 3y ago

It seems that if you want to murder someone and get away with it you should do it in a car.

LatteLazy 3y ago

On the other hand, what's the point in locking up someone who (a) statistically will never do it again and (b) never intended or foresaw that this would happen?

andrewfurey2003 3y ago
I suppose people would take getting into and driving a car more seriously
- LatteLazy 3y ago
  Would they? A lethal accident is something like once per million journeys. I don't think about that risk when I start my car. Adding a prison sentence would still not make me think about it, because it's such a tiny risk... Sorry if that's a bit blunt. It would definitely be better if people were more safety aware. Maybe making people watch videos like this as part of (re)training of drivers would be more effective?
  - badcppdev 3y ago
    If you knew that the penalty for killing someone while speeding was 10 years in prison then would you stay under the speed limit?
    - LatteLazy 3y ago
      No, No more than I currently do anyway. I already strongly don't want to kill anyone. But I also know the chances of me doing so are very very low. So that 10yrs is very heavily discounted by the tiny probability of it happening. I guess an analogy for this is that I don't change my behaviour after I buy a lottery ticket. Sure it is not technically impossible that I will be a millionaire soon. But I don't take the precaution of lining up a lawyer and a tax accountant because the chances of it coming to pass are so tiny...

NoZebra120vClip 3y ago

Imagine what the penalty would have been if the "backup driver" didn't already have a felony conviction record.

spoonjim 3y ago

Killing someone unlawfully should have a minimum of 10 Christmases in prison.

Lolaccount 3y ago
A 10 day sentence is pretty weak ...

friend_and_foe 3y ago

I think youre drawing the conclusions you want from the scenario. The truth is that Uber removed safety features from the car that could've stopped this and the car only recognized pedestrians if there was a crosswalk. If you ask me this is not the woman's fault, they're framing it like she was "at the wheel" because they want to shirk responsibility, and probably paying good PR money to these news rags to keep the narrative going too. Uber did this, they knowingly put an unsafe machine on the road and it killed someone. They put someone at a computer in a call center as a fall guy and it worked.

buro9 3y ago
I didn't draw conclusions about this specific instance. Consequences should exist, and should be a deterrence against unsafe behaviour no matter the party. Consequences for corporations should be more significant than for individuals IMHO. If Uber contributed to this in their training of their safety driver, the equipment in use, disabling built-in safety features, and so forth... then they should have penalties severe enough to adjust their behaviour too. My thinking is that the more vulnerable a road user, the greater the ownership is on the less vulnerable, meaning the greater the consequences of their choices. What I didn't do is limit "their" to just a driver / safety driver.

goodpoint 3y ago

"consequences" do not solve the problem because people cannot self-evaluate their driving skills and general attention level. In US getting a driving license is shockingly easy and cars are huge and inherently dangerous. You need train and trams.

unsupp0rted 3y ago

I don't see what making the woman do hard time for 10 years would accomplish. People kill each other with vehicles all the time, because apes that attempt to drive 2-ton metal canisters at 100 kph around other apes doing the same, and some apes on bicycles or walking by, will inevitably do something catastrophic. I don't really blame the apes at the wheel, since they're trapped in a system that is much larger than them. After a year or two of intentional driving experience leading to muscle memory, ape brains are wired to fall into comfortable inattention in these scenarios. What can you do, other than give up driving and give up walking where cars go? We weren't designed to do this safely.

nullc 3y ago

> In the USA alone, that's 117 people per day. Against how many passenger miles or hours?

acumenical 3y ago

Generally I agree with you, drivers get away with murder too easily, however in this case this woman's crime was not paying attention in a self driving car. What if instead she had been paying attention but applied the brakes too late, but it wasn't enough? What if she was paying attention but was confused by instructions from her employer to let the AI do its job? What if she had been in the same situation before but the car had previously stopped in exactly the same circumstances? Even engineers become confused by LLM and think that it's so quirky that ChatGPT "hallucinates" when that's the nature of the technology. So how would you expect a layperson to actually know when they should apply the brake pedal in a moving vehicle that's supposed to drive itself (maybe)? This is a bit more complicated than driving drunk and I'm fairly certain that the company responsible for the death did not seek a safety expert to copilot the vehicle, but rather any warm body willing to take an easy dead-end job.

darkclouds 3y ago
> This is a bit more complicated than driving drunk I'd argue no, manganese needs to be quantified in the blood stream when drunk driving imo, the state doesnt do that, but might if the doctor is called to take a blood sample having refused a breath test at the road side and/or in the police station. Manganese has a half life of 10seconds in the blood stream, and alcohol is very good at releasing it into the blood stream for elimination and thus deficiency. When is a drunk person or a baby or toddler, not suffering from manganism which can be likened to Parkinsons? Plus there are some gut bacteria which will produce ethanol from sugars, which is why some people always blow positive for drink driving even when they have not been drinking. Simply measuring alcohol is not enough, and if you happen to be able to blow you own trumpet, then many musicians will know that they can do a rebreathing trick, which means they will never fail a breath test, even if they fall out of the vehicle!