Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

San Francisco cops pull over a Cruise driverless car for no lights on

theverge.com

Post reply on HN
Print view View original

161–170 of 504 posts

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#161

Post by CogitoCogito »

MathCodeLove wrote: post #153
sseagull wrote: post #131

Earlier quoted context omitted.

I generally am in agreement. However there's a complication in my mind. My dad was a police officer and I grew up around police. It's there's one thing to know about police, it is that they are human, and subject to all the emotions, inconsistencies, and foibles as other humans. However, they are given an extraordinary amount of power, with very little accountability. Think about what it would take for a regular pers…

Alright but again, the anecdote of "This one time in Arkansas" can't reasonably be used as a justification for stating "All police are bad and out of control all over the country"

> Alright but again, the anecdote of "This one time in Arkansas" can't reasonably be used as a justification for stating "All police are bad and out of control all over the country"

In this post https://news.ycombinator.com/item?id=30987681 you wrote:

> ...There may be 1 in 100 truly awful police men and women out there, just as their are truly awful people in every known profession.

Your complaint about others' anecdotes would hold more weight if you weren't just making up statistics.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#162

Post by eru »

bee_rider wrote: post #154
hakuseki wrote: post #83

Earlier quoted context omitted.

It does sound silly at first, but I think suspending the one car is actually a rather elegant solution on several counts. 1. Easy to implement 2. Doesn't punish fleet scale (compare: suspending the whole fleet means a fleet that's twice as large would be suspended twice as often) 3. Punishes the fleet owner in proportion to the frequency of the error 4. Is exactly correct when the cause is really a hardware problem s…

The whole fleet should be grounded until the issue is debugged. Without debugging the issue * It might be a software issue (maybe the cars self-destruct when they encounter a solar eclipse -- silly example, but if the cars malfunction on events which are rare but wide-spread and correlated this could cause an emergency). * If it is a hardware issue, it must be understood how a hardware failure can put the device in a…

> * It is good that the fleet is punished disproportionately to the frequency of errors. We shouldn't accept a linear disincentive for putting dangerous vehicles on the road.

We already do..

Putting driverless cars to a much higher standard, makes perfect the enemy of the better. Human drivers are not all that great.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#163

Post by MathCodeLove »

jcranmer wrote: post #155
MathCodeLove wrote: post #126

Earlier quoted context omitted.

As for point 2. Okay, but now cars do exist. Do you think it's reasonable for cars capable of traveling 40mph should be stuck behind pedestrians at walking speed? Our modern transportation infrastructure exists for a reason, and that reason is speed/effeciency. Most areas with heavy pedestrian traffic have sidewalks. If I'm going 60mph down the highway then the presumption that I should be subjected to pedestrians ca…

Already today, cars do not travel at 40mph in those areas with heavy pedestrian traffic--the typical speed limit on a city street is 25mph, and even that is probably above typical speeds. If you want cars to travel more quickly, you need to resort to highways that completely cut off pedestrian access, which has seriously deleterious effects on the neighborhoods it travels through. If you try to compromise, you end up…

Alright, but 25mph is still significantly faster than walking speed. I do think that mass transportation would be nice and we could use a better public trans infra in the US, but that doesn't make the fact that waking is still much less efficient that driving. Even with cars taking up much more space, a bunch of cars traveling at 25 mph are able to carry a many more people through a distance X in a period of time T than the equivalent walking would be able to. The alternative that everyone seems to be forgetting is having roads for cars and sidewalks for people, which already exists in the vast majority of areas with a high potentiality for pedestrian traffic.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#164

Post by eru »

logifail wrote: post #144
snovv_crash wrote: post #135

Earlier quoted context omitted.

It depends. If it's an issue with just that vehicle then yes, you don't ground the whole fleet. If it's non-safety-critical, then it needs to be fixed asap but nothing gets grounded. If it's safety critical, then yes, all planes are grounded. But, I'd argue FAA levels of safety aren't suitable for cars. If a car has an engine failure or even a structural failure, they can usually just stop. Planes don't have that opt…

> I'd argue FAA levels of safety aren't suitable for cars. If a car has an engine failure or even a structural failure, they can usually just stop. Planes don't have that option. I had supposed we were thinking primarily about software faults and specifically those relating to the self-driving bit of a self-driving car. Best case: self-driving car gets lost and is found with flat battery miles from "home" (?) Worst-c…

I wonder whether that would just lead to people using slightly tweaked variants of the same algorithm branded to be different?

(Of course, you don't want to fork too much, because there's probably a certification per algorithm needed. But it might make sense for Google to officially field a thousand 'different' algorithms, so that a single incident doesn't ground the entire fleet?)

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#165

Post by Steltek »

MathCodeLove wrote: post #106
MereInterest wrote: post #90

Earlier quoted context omitted.

People did say exactly the same for regular cars. The result was a massive advertising campaign to establish the concept of jaywalking, and to make the street the sole domain of the car. I can easily imagine a similar campaign to label pedestrians as irresponsible for walking outside without an ultrasonic emitter, or a lidar reflector, or whatever integrates with the self-driving system. Whatever shifts the responsib…

You say "advertising campaign" as if it were some conspiracy by the world's big auto manufacturers 1. Do you have any sources? 2. You don't think that roads should be a car only domain? What do you think roads are for?

> 2. You don't think that roads should be a car only domain? What do you think roads are for?

You've clearly never had a UPS driver scream "Use the bike lane!" at you after parking his truck in that same bike lane. Drivers have bizarre reflexes for how to use roads.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#166

Post by eru »

AvocadoPanic wrote: post #105
hakuseki wrote: post #94

Earlier quoted context omitted.

Yes, and that's why suspending the one car might seem counterintuitive. But that doesn't mean it's a bad solution.

Can we have lawn darts back if we suspend the darts that maim or kill?

You can't even have Kinder Surprise..

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#167

Post by MathCodeLove »

Cockbrand wrote: post #122
MathCodeLove wrote: post #106

Earlier quoted context omitted.

You say "advertising campaign" as if it were some conspiracy by the world's big auto manufacturers 1. Do you have any sources? 2. You don't think that roads should be a car only domain? What do you think roads are for?

1: e.g. https://en.wikipedia.org/wiki/Jaywalking#Origin_of_the_term 2: Roads are also for bicyclists, motorcyclists, and even pedestrians. It's true though that car drivers often think that roads are a car only domain and behave aggressively towards the other mentioned groups

2. No, sidewalks and bike lanes are for pedestrians and bicyclists. Bike lanes are less common than sidewalks so I think it's reasonable to include them on low-speed/highly congested stretches of road, but to assume they should have a full access to, e.g. and interstate or highway is ridiculous and would hinder literally thousands of people.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#168

Post by MathCodeLove »

CogitoCogito wrote: post #161
MathCodeLove wrote: post #153

Earlier quoted context omitted.

Alright but again, the anecdote of "This one time in Arkansas" can't reasonably be used as a justification for stating "All police are bad and out of control all over the country"

> Alright but again, the anecdote of "This one time in Arkansas" can't reasonably be used as a justification for stating "All police are bad and out of control all over the country" In this post https://news.ycombinator.com/item?id=30987681 you wrote: > ...There may be 1 in 100 truly awful police men and women out there, just as their are truly awful people in every known profession. Your complaint about others' anec…

I think it's extremely clear that I'm not trying to portray that exact number as a fact and more as a general portrayment of "most police officers are good, just as most people are good"

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#169

Post by throwawaylinux »

coldcode wrote: post #115

So what would happen if the passenger is arrested by the police for being in a car that refused to stop? I wonder what legal principle could be used to justify this. Involuntary failure to stop? Are there any laws about driverless vehicles and who is responsible for problems?

> So what would happen if the passenger is arrested by the police for being in a car that refused to stop?

Just for being in the car? Presumably they would be questioned and released.

> I wonder what legal principle could be used to justify this.

"Probable cause" of having committed a crime is typically the principle I believe.

> Involuntary failure to stop? Are there any laws about driverless vehicles and who is responsible for problems?

I don't know and can't imagine what involuntary failure to stop might be. I'm sure they could fit an existing law to the situation if they suspected the passenger was instructing the car to not stop.

Re: San Francisco cops pull over a Cruise driverless car for no lights on

#170

Post by eru »

qayxc wrote: post #81
pxmpxm wrote: post #74

Earlier quoted context omitted.

How is it non-deterministic? If you run the same trained model on the same input data, you should arrive at the exactly identical output.

> How is it non-deterministic? It's non-deterministic in a practical sense; you cannot reproduce the exact circumstances that caused a certain behaviour. Maybe a bug flew by one of the cameras at a certain moment, or a pedestrian with certain type of clothing passed by. You get the idea.

Depends. In principle you can log exactly what data the camera sends to the decision making unit. And you could replay that data.

The camera itself might not be deterministic enough to reproduce the same data with the same beetle, but replaying the log should work.

(Keep the log in a 'black box' on the car, and perhaps only keep the last x hours of data, if necessary.)

I think Waymo already does something like these kinds of replays.

Post reply on HN