I know I posted about this yesterday, but this article does a much better job than I can.
You must log in or register to comment.
This isn’t even political and quite likely the first article posted in this community with which I agree.
I’m against this because it’s a terrible idea in terms of privacy and security.
It requires the kill switch to be triggered by software and I’ve worked in software… there will absolutely be bugs, hacks and/or abuse, possibly all of the above.
And no, that’s not to say I’m against all software, but something that restricts or cuts power to your car seems like a terrible fucking idea.
I might be driving too erratically for whatever baseline they code, for an actual good reason… maybe I’m in a shitty backwoods road with holes bigger than the car.
Maybe the sensor is just defective.
People also drive non off-road vehicules… in off-road conditions.
I’ve rescued a wounded man with an internally bleeding leg, crushed by machinery, way out in the sticks, at the far edge of a field and several shittier backroads.
The ride back to civilization was an absolute shitshow and we barely made it back in time to save his life and his leg already. Interfering from some shit algorithm made with a perfect road in mind would have killed him. Which reminds me I should call them and visit them for a hug.I don’t ever want this shit in my car and I’ve never driven drunk or impaired, can’t even remember my last speeding ticket from decades ago.
Obscure, proprietary, closed source software blobs isn’t a solution.
The question of whether government should be able to disable your vehicle is indeed a political one.
It’s not the government shutting down your vehicle. It’s a regulatory requirement for car manufacturers to include a new safety device that the government is creating a standard for.
And disabling is not a requirement of this regulation. It says the device should disable OR limit operation of the vehicle when impairment is detected.
None of that hair splitting changes the basic fact that this is a question of the government taking away people’s car. The fact that it might only slow you down, the fact they’re ordering others to carry it out instead of doing it themselves, it’s all the same thing.
Do you feel the same way about seatbelt laws?
Do you feel like it’s a bad thing when the government takes the license (and therefore car) of a drunk driver?
It requires the kill switch to be triggered by software and I’ve worked in software… there will absolutely be bugs, hacks and/or abuse, possibly all of the above.
As have I, and specifically train software. It’s a bitch to work with but not impossible to create a safe product.
And no, that’s not to say I’m against all software, but something that restricts or cuts power to your car seems like a terrible fucking idea.
The language of the bill states that “identifies whether they may be impaired and prevents or limits motor vehicle operation”, so a full power cut isn’t a requirement. Like I said in my other comment, a speed limit might be the end result of this (at least to me it makes the most sense).
I might be driving too erratically for whatever baseline they code, for an actual good reason… maybe I’m in a shitty backwoods road with holes bigger than the car.
So there is a difference between a shitty bumpy road an erratic driving. Even the most grotesquely maintained path will have the majority of the jostling kept to the vertical and roll axis. The constant vertical change and rotational change would make it pretty clear it’s just a terrible road, so a system only really looking at the yaw axis wouldn’t care. And with some other sensors to provide context, such as one to measure the length the vehicle’s shocks are at, could rule out bad roads.
I have a device in my car from my car insurance company to monitor my driving. I don’t go on shitty roads much, but when it detects me making too sharp of a turn, or too sudden braking/acceleration, usually it’s justified. If I make them a whole lot in a short period of time, then it would be justified for the car to self-speed limit.
Maybe the sensor is just defective.
This is probably the worst case scenario, but not entirely the end of the world if the system has a backup, and an error code for one of the sensors going out.
Obscure, proprietary, closed source software blobs isn’t a solution.
I agree with you on this. It should be open source and well documented.
I have a device in my car from my car insurance company to monitor my driving
Lmao, checks out that you’re just welcoming in the surveillance stare with open arms
My car company is not a part of the state, and I am cheap ass. I am a good driver and I want my insurance rates to reflect that.
I get good insurance rates, and I told them to stuff it when they “offered” to spy on me wherever i go.
And lmao that you actually think “good driving” and “driving favored by insurance actuaries” are even remotely the same.
you actually think “good driving” and “driving favored by insurance actuaries” are even remotely the same.
I never said they are the same. But there is enough overlap to get me a decent discount.
Unless you just had your rates in the absolute shitter beforehand, you have very low standards for what a decent discount is, especially when you consider that you’re normalizing them taking an overwhelming amount of data about you to create future insurance profiles for everyone.
Oh no!
Even the most grotesquely maintained path will have the majority of the jostling kept to the vertical axis.
If you’ve ever driven out in a field or in the boonsticks, that’s just not always the case.
In some places, you simply have to swerve to avoid the worst of it because otherwise your tires and/or direction just won’t survive. Like, you’re not bouncing up and down through potholes, you’re dodging ditches, roots, pits, divots, rocks.
Sure you could drive 10kph, but sometimes it’s not that good of an idea, like say, in an emergency with your passenger bleeding out.
Is it somewhat of an exception? Sure, but something that you can’t disable and takes some measure of control away from you isn’t something I’m excited about.
In theory, that could be nice in some cases, but that seems dangerous.Like, you’re not bouncing up and down through potholes, you’re dodging ditches, roots, pits, divots, rocks.
If you’re having to avoid that, it sounds like you were either already going slow, or were going straight over those obstacles. In either case a good system would be able to tell you’re going a low speed and it’s safe, or it’s just a bumpy road.
Is it somewhat of an exception? Sure, but something that you can’t disable and takes some measure of control away from you isn’t something I’m excited about
What about a comprimise of vehicles containing such a device being subsidized? You have the choice, and a safer car is made to be the better more competitive choice. But for those who really worry about an edge case they still have an option.
If you’re having to avoid that, it sounds like you were either already going slow, or were going straight over those obstacles
No to both.
You seem stuck on the vertically of this scenario, and a trained software would likely make the same assumptions, yet they’d be wrong in some cases.I did specifically mention swerving and not merely bumpy, from an actual real life scenario that I remember quite vividly.
I wasn’t driving over any of these obstacles, because the car would have simply been destroyed, we’d have crashed and both died.
I wasn’t driving that slow either, because my passenger was bleeding the fuck out.
The doc said it was a matter of minutes.Software works fine for things where the driver’s intent can be determined more clearly, like traction assist. I got no issue with that.
In this limit/kill-switch, driver intent cannot always be determined reliably because some factors depend on things there cannot be a sensor for.I fail to see what problem this solution is supposed to address other than giving more power and data to companies and governments. Odds of this being a transparent, properly audited, open source solution are nil.
If it merely flagged you for review or something, maybe?
I’d have no issue with such a system if it merely tried to wake the driver up when it thinks you’ve fallen asleep.
If it takes control away from you, possibly in some fringe case emergency that’s not accounted for in whatever software, it can fuck off.Drunk drivers would either not opt-in, or bypass them illegally… like they already drive drunk, without licenses or plates illegally.
In Canada, we already have alcotest machines mandated in cars for people that have DUI infractions, maybe this technology could complement that.
For the general populace, it’s not something I’d be excited about.You seem stuck on the vertically of this scenario, and a trained software would likely make the same assumptions, yet they’d be wrong in some cases.
Fair enough.
I fail to see what problem this solution is supposed to address
Drunk driving
Odds of this being a transparent, properly audited, open source solution are nil.
I don’t disagree with you there, but just because it might end up being closed source and such doesn’t mean it will be connected to the internet and giving your info to companies/the government. The bill specifies that it should be a passive system.
If it merely flagged you for review or something, maybe? I’d have no issue with such a system if it merely tried to wake the driver up when it thinks you’ve fallen asleep.
I wouldn’t have a problem with either of those two options.
Drunk drivers would either not opt-in, or bypass them illegally… like they already drive drunk, without licenses or plates illegally.
I don’t think it is that clear cut. Drunk driving isn’t exactly a premeditated crime. It’s one that generally happens on a whim. Sure, some people know they have a habit of doing it and would find a way to opt out, but it would still help address other drunk drivers.
In Canada, we already have alcotest machines mandated in cars for people that have DUI infractions, maybe this technology could complement that.
I would be open to that as well.
If we are really being honest, the easiest solution is to stop having car dependent cities and transportation networks. Nobody is gonna drive drunk when the train/subway/bus/sidewalk can get you home safely, quickly, cheaply, and easily.
Take a look at Japan’s drunk driving fatalities. They have almost none. Part of that is their ‘draconian’ laws/low legal alcohol limits, and also probably culture. But they have fantastic public transportation, and it’s a great option for drunks to get home.
Bad roads is my first thought too. The roads in my state are horrible. I tried one of the progressive “snapshot” devices when I was younger, and that thing acted like I was in an accident a few times juat because of potholes and bad pavement joints. So many states would have disabled cars on the road if they are too sensitive with their little kill switch.
A decently designed device would be able to account for bad roads by comparing vibration in the X and Y axis versus the yaw and roll of the vehicle. Progressive’s device isn’t that.
If the system they end up designing is capable of accurately telling the difference, then it seems to me to be a good safety feature. Otherwise, sure, it is a terrible idea.
The question being presented isn’t “Is it possible to design a solution?” The question is “Will the corporate stooges involved bother to implement that solution?” As shown by the Progressive doodad, they probably won’t.
Corporations won’t. But the proposed system is being designed by the government.
The same government that said “Big vehicles can pollute more” which caused our SUV and big truck epidemic? That government? You’ll have to forgive me for thinking they will just have the auto industry design it for them, like they have with laws surrounding fuel mileage and pollution standards.
Would you be saying this in response to regulation requiring seatbelts?
Does a seat belt allow someone to remotely disable the car? Try a different strawman.
This isn’t even political and quite likely the first article posted in this community with which I agree
How so is it not political? The government forcing your car to shut down seems like the definition of political.
And given that you don’t support the government doing this, is that not also political?
Yea I meant something else at the time, but fumbled it.
Probably meant something like it shouldn’t be partisan.That checks out. Calling it not political seemed really unusual, but it seems there’s people who just spam “this isn’t political” under every post, so I wasn’t really sure.
No I probably just had a brain fart when I wrote that because it makes no sense to call in non political.
The term “kill switch” is trying to get you to think that the police get a button to turn off the car, which is the one thing this law doesn’t do. It’s a mandate for the government to devise a passive, local system within your car to detect erratic driving.
Here is a more detailed explanation of the bipartisan bill in question:
Specifically, Section 24220 of the bill directs the National Highway Traffic Safety Administration to develop rules that would require new cars to be equipped with technology that “passively monitors the performance of a driver,” identifies whether they may be impaired and prevents or limits motor vehicle operation “if an impairment is detected.”
The agency has three years to finalize the standards for which technologies cars should use, according to the bill. Automakers then have between two and three years to implement those standards. The earliest implementation date for the technology is 2026.
When laws were introduced to require seat belts in cars, with a legal mandate to use them while driving, people declared it an infringement of their rights.
https://www.businessinsider.com/when-americans-went-to-war-against-seat-belts-2020-5
When laws were introduced to toughen up drunk driving laws, people declared it communism.
https://www.youtube.com/watch?v=xSDniOoR3Lw
This is nothing new. Regulation should uphold public safety, and passive safety systems make sense in the age of information.
If I had to take a guess what the end result will be, it will probably just be a gyroscope that will disable the car if too much swerving/erratic behavior is detected. But because you can’t just stop a car in the middle of the freeway, it will probably set an internal speed limit to the car just like a train does. Probably a speed low enough for you to get off the road but not high enough for you to hurt somebody even if you’re drunk.
But all of that is nuance, and you don’t often get that around here.
But all of that is nuance, and you don’t often get that around here.
All the nuance in the world doesn’t matter when it’s just different flavors of shit.
Thank you for your high effort contribution to the discussion. Really helps.
I mean there’s not much to say. You’re basically just trying to explain that you didn’t serve me a plate of human shit, it’s actually dog shit, from only purebred poodles. Cool distinction, I still don’t want to eat shit.
I didn’t serve you anything. And those who are serving you, are not serving you shit. They’re serving you public safety.
I didn’t serve you anything
Yeah, sorry, I forgot you’re just here to run defense for all the garbage the democrats want to shovel onto us. My bad
The problem with this system is the assumption that roads will be smooth and not rough… the last thing you need is the car to decide to power down on a mountain road with lots of blind spots, or somewhere that’s dangerous to suddenly be slowing down well below traffic flow.
the last thing you need is the car to decide to power down on a mountain road with lots of blind spots
The language of the bill doesn’t specify it requires a total power down.
somewhere that’s dangerous to suddenly be slowing down well below traffic flow.
If I were the engineer designed to create this system, I would probably take a page out of a train operator’s book. If you fail to drive safely, it will give you an unignorable alarm as a warning, and if the issue persists it will automatically slow the train down to an eventual stop. Only with the car, it would slowly ramp down the vehicle’s max speed to something like 5mph, enough to get off the road but not enough to hurt people.
How do you know 5 mph is fast enough to get out of harm’s way?
If I were designing it, it wouldn’t be instantaneous. It would ramp down. So it wouldn’t be much different than your engine giving out on you, which roads are already designed to handle.
Except when the road is a shoulderless corridor, and now you’re blocking up traffic. Worse still if it’s caused by faulty sensors and not by the driver.
I would rather have traffic slightly blocked up than a drunk driver kill people.
And as for the sensor thing, it’s not that big of an issue. Trains have 3 speed sensors. If one fails, you get a warning. In a car system such as this, the check engine light should turn on until the problem is resolved. If a 2nd/3rd goes out, then the car shouldn’t start, but it will allow you to finish an existing trip.
If I were the engineer on this project, that’s how I would start, I’d literally rip off how train safety systems work and tweak it to make sense on the road.
I was watching dashcam / accident videos and I’d seen one where a car got disabled in the middle of a highway underpass, which was a blind crest. Car after car after car would enter the underpass, unknowing of the disabled car there, and make emergency maneuvers, barely missing the car every time (and some didn’t). Several vehicles crashed at full highway speed into either side of the underpass trying to avoid the disabled car.
That doesn’t sound like safety to me.
As for trains, remember that trains only need to keep track of one axis of motion, the forward axis, as the train is not going to be maneuvering, as that would cause it to derail if it even could. Maximum speeds are a known factor on the line, and accurate timing is important to coordinate multiple trains on a line. Cars aren’t trains, and the requirements are a lot less predictable.
Will the software know the difference between winter weather and drunk driving?
Will the software know the difference between a windy mountain road and drunk driving? How about drunk driving on a windy mountain winter road?
How about fix the problem of drunk driving being so prevalent, instead of adding additional points of failure to an already overly expensive vehicle?
That doesn’t sound like safety to me.
Sounds to me like the real issue there was the design of the road being such that it didn’t have visibility.
As for trains, remember that trains only need to keep track of one axis of motion
Technically that’s not true, they keep track of way more than that. But I understand what you’re getting at, and it doesn’t quite hit what I’m getting at.
I’m saying we have a pretty good idea of what goes into a safety system on incredibly dangerous vehicles (trains), and some of it can be transferred to another set of incredibly dangerous vehicles (cars).
Will the software know the difference between winter weather and drunk driving?
If you’re swerving like you’re drunk in winter weather, are you really arguing that you should be allowed to keep driving? If you’re swerving at high speed in winter weather, you’re already doing something wrong and going to get somebody killed.
It’s just one more thing to break and adds to the cost of a car.
Cars are overly expensive now. Imagine as they keep adding garbage features.
Car are in general a terrible way to get people where they need to be, and cost is part of that. It isn’t cheap to design/manufacturer a safe car.
Trains on the other hand is where it’s at.
The problem with this system is treating adults like children.
deleted by creator
I wouldn’t worry too much. As soon as these kill switch cars become available to the public, someone will figure out a way to disable it.
And the 2025-old used car market will become hot
