These seems to just confirm there point while using a lot of weasel words to say it's not a "kill switch"....?

Seriously I've not read such a round about stretch of a justification in a long time.

It's needlessly wordy, that I agree on, but here's the important part: the actual text of the bill is "equipped with advanced drunk and impaired driving prevention technology"

I guess one way you could implement that is a remote kill switch, but at least when I first read the text I thought it would just be something like a built-in breathalyzer you had to pass before the car would let you drive.

People claiming this is evidence of a kill switch are hard extrapolating with (as far as I've seen) no evidence that gets you from the bill's text to a kill switch.

There is no such thing as a switch that can only work from some arbitrary cause. The switch is either capable of disabling the car or not.

"equipped with advanced drunk and impaired driving prevention technology"

Is indistinguishable from

"equipped with driving prevention technology"

"equipped with a mechanism that prevents driving without the key"

is distinguishable from

"equipped with a mechanism that prevents driving".

Doesn't the reason you can't drive actually matter?

The ignition lock doesn't know or care what the reason is that you don't have the key.

On the books, the lock is there to allow only the owner to operate the car.

Yet here you are, without the key, but still the owner, and yet you still can't operate the car.

The ignition is not preventing you from operating the car because you are not the owner.

Tools get misused constantly and there's no end to mission creep when "think of the children", "you're helping the terrorists win", "the cartels were involved", etc. cards are played. Also we are in an era of incredibly flexible definitions of words. What you consider to be 'impaired driving' today might not be what the government considers it to be when they don't like you.

By this logic I can stretch any law to any meaning.

The original claim was a bill mandating a remote kill switch and we've now backpedalled to "words might have a different meaning tomorrow".

So all laws bad, or just this one?

I almost agree with you.

The way Snopes presents it sounds like there is enough wiggle room in the requirements for it to be a kill switch:

> Passively monitor the performance of a driver to accurately identify whether they are impaired.

> Prevent or limit operation if impairment is detected.

> “Passively” detect whether the BAC of a driver is equal to or higher than the legal limit. In such cases, the system could “prevent or limit motor vehicle operation if an impairment is detected.”

What????

Cars already have this! The base model Corolla has had this for half a decade! Hyundai, MB, Nissan, all have this across the board.

If you're impaired for any reason (but mainly assured to be drowsiness), it detects your behavior and gives you a warning:

https://www.autoguide.com/blog/wp-content/uploads/2014/05/ti...

https://di-uploads-pod5.s3.amazonaws.com/sorgnissan/uploads/...

https://www.autotrader.com/wp-content/uploads/2020/08/RT_Dri...

All the bill does is say now those systems should escalate beyond a warning and be able to stop the vehicle.

These cars already have all the info they need, if you "accidentally" drive like an intoxicated person, it's going to be for a reason you need to be stopped over!

Like a medical episode, dangerous lack of sleep, texting and driving.

It has literally nothing to do with being a remote kill switch.

Saying it's a remote kill switch is like saying the mandatory backup camera system is as a mandatory surveillance tool...

It's such a huge leap you can't possibly make it in good faith...

What if you have to drive like a maniac to get your friend to the ER before he bleeds out?

(Yes, a friend of mine told me how he had to do just that. His friend died anyway, but it wasn't because the car quit.)

There are lots of cases where I might need to get the most out of the car.

Or maybe I just want to practice parallel parking. Erratic back and forth driving and turning.

So the car according to some algorithm decides to shut down the car and strand me. The worst part is, I have no idea what criteria that algorithm is using, so I have no idea what line to not cross.

None of this and I mean _none_ would trigger it.

Not recommending this or proud/bragging/endorsing this, but I drive fast. Too fast. 100 MPH for me is what most people feel doing 5 over.

I have *never* triggered these alerts unless I was genuinely tired.

-

You're talking about a system you don't have the faintest clue about, instead of whipping out irrelevant fantasy situations, ask how it works???

It will *not* react to intentional aggressive movement, it reacts to clearly unintentionally movement that are consistently late and low steering torque.

This is not stuff you can replicate "practicing parallel parking", or driving like you're re-enacting fast and furious!

I expect better of HN, you're jumping from a misleading interpretation to FUD without taking a moment to learn.

> a system you don't have the faintest clue about

Which I explicitly stated:

"The worst part is, I have no idea what criteria that algorithm is using"

> clearly unintentionally movement

How does a machine know I didn't intend it?

You know... when most people have no idea what they're talking about they seek clarity instead of writing 4 paragraphs of fantasy.

But you do you.

These systems don't trigger for "potentially bad" behavior like speeding, they trigger when you drive like an intoxicated or impaired driver.

Intentionally acting like an impaired drive is like intentionally walking like you're drunk... gravity doesn't care if you were pretending to stumble, you're still coming down.

So it doesn't matter if you intended to veer in and out if your in your lane with barely any steering torque being applied... that's how you cause an accident.

If you were dodging obstacles you'd be applying sharper input than a drunk person. You can't "fake" drunk without being a danger to yourself and others.

How does it know you're wandering in and out of your lane?

You can keep trying to go for gotchas... but it's not worth your time or mine.

What do you think is going to happen, ADAS is going to have a tracking error that makes you look like an impaired driver?

These systems are even starting to take in eyesight and

-

You're like a person going in front of a judge trying to argue:

"Well the officer heard gunshots and watched me, the only person in the room, run away with a smoking gun as the guy fell..."

"But maybe I didn't shoot him!! Maybe I was shooting out the window and felt like going for a run!!"

You're not really going to fool anyone... if it drives like it's impaired, it's impaired. Doesn't matter if it's alcohol, sleep, texting, etc.

Yeah, well, I've been in software for 45 years. Magical algorithms like that always have both bugs and unintended consequences.

"if it drives like it's impaired, it's impaired"

It's just magic I'm supposed to trust.

No thanks.

P.S. I worked on the 757 stabilizer trim design. While I didn't work on the avionics boxes that fitted to it, I know was designed to never allow the software to have the last word. Because software has bugs in it. It always does.

Yes, I know about the 737MAX. There's a switch to turn physically off the stab trim. All the electronics that drive the trim, including MCAS, are upstream of that switch, and that switch rules. Another switch that overrides all the software is the column trim switch. The 3rd crew that experienced MCAS failure that you never hear about used these switches to override MCAS and land safely.

P.P.S. I let my car drift out of the lane if there's no oncoming traffic to give a crazy cyclist / kid on a skateboard more room, or to give that mom with kids more room, or to avoid an icy patch, or to avoid a pothole, or to get my wheels on a drier stretch of pavement to avoid hydroplaning. Heck, if it's safe to do so, I'll move over a full lane to avoid kids.

I've a very soft driving style, because I'm used to cars with very direct steering, and when I get into rentals these systems trigger like crazy with me, they're super annoying.

"Driving like a maniac" to save one person is essentially one possible answer to the trolley problem.

I'm not saying that I condone that particular answer being applied to this particular case, but it's well without the bounds of moral philosophy to say "you should not be able to make this choice".

If the remote kill switch exists, it's going to be hijacked into government having the power to disable your car. It starts innocuously, something that everyone will agree with, like drunk drivers. Next will be drug dealers, who could disagree with that? Then deadbeat parents who owe child support. Then tax-owing deadbeats. Then who knows what other classes of "undesirables".

Proof? It already happened: driver licenses already went through every one of those steps, getting hijacked way beyond the original purpose of denoting that you had the training to drive a car, into government-imposed threats against all sorts of "bad" behavior.

I don't trust any expansion of government power in good faith.

Again, this is just like saying that a camera existing enables the government to hack it, so rear view cameras are part of forming a surveillance state.

There is literally no reason for automakers to do this, and yes I know now the FUD comes in "well they will cuz reasons! cuz gubbmint!"

But just like the rear view camera... you're imagining that they'll spend significant time and money to do something with no use other than incriminating themselves...

So sure maybe your rear view camera can be rigged to stream to their servers 24/7, some already record and upload incidents anyways right?

But it's a massive jump from the happy path.

This is 100000x easier to do on the car without looping in a remote server and opens them up to 100000000000x more liability if the interface is hacked.

They wouldn't do it.

> Again, this is just like saying that a camera existing enables the government to hack it, so rear view cameras are part of forming a surveillance state.

That might be the best example of a straw-man that I've ever seen.

>> Again, this is just like saying that a camera existing enables the government to hack it, so rear view cameras are part of forming a surveillance state.

>That might be the best example of a straw-man that I've ever seen.

I think that was the point.

Replace "rear view cameras" with "onboard tools to detect and deter/prevent impaired driving" and you have the same straw man.

This right here, that was an ironic comment to read on their part...

Why do you assume these systems are static? It's not a given that the systems available in 2025 will have any resemblance to the ones available today. And in fact they may have mandatory upgrade abilities, so the car you buy may not be the same as the car you have 5 years later.

Its a control agenda.

The eco-agenda is the provided pretext that will mean we willingly accept what is offered, in the name of saving the planet.

"a control agenda" is one (landmine) label meant to represent a set of complex interactions; this kind of drastic oversimplification is daily fodder for politics, but literally destructive to actual fact-based inquiry leading to effective policy. Similar arguments have been made regarding product safety and hygiene requirements by "government" !

Except that such all-or-nothing logic is completely appropriate when talking about security of computational domains. Taking a device that is supposedly mine and configuring it to work against my interests repudiates the idea that I am the owner of it. Rather than representing my interests, it is now an agent of someone else's top-down policy. And if enough people modify "their" devices to remove such restrictions, then there's a good chance that even more systems of control get deployed to prevent them from doing so.

I was under the impression people are buying Teslas voluntarily, not because it’s mandated by the government.

I don't see how that's relevant. Both government and corporations are independently capable of pushing top-down controls onto society.

>Except that such all-or-nothing logic is completely appropriate when talking about security of computational domains. Taking a device that is supposedly mine and configuring it to work against my interests repudiates the idea that I am the owner of it.

You mean like smartphones, proprietary operating systems, streaming devices, many (most?) IOT devices, digital books/music/movies and a host of other things?

Sorry, until there's some law/regulation (e.g., broadly confirming the First Sale doctrine[0] and/or similar stuff) that ship has sailed, my friend.

[0] https://en.wikipedia.org/wiki/First-sale_doctrine

Mass adoption of centrally controlled devices doesn't change the paradigm of reasoning about computing freedom, though. This trend just means that the majority of computing devices aren't under the control of their owners, not that the idea of computational freedom is irrelevant.

In fact I would say barring some very strong laws that make de facto locked down devices outright illegal, there will always be more proprietary devices than libre ones. One of the chief goals of locking down computers is decommodification, meaning taking what could be a general purpose device and restricting it to one specific purpose. Even caring heavily about software freedom - I've got two librebooted desktops, two me_cleaned laptops, one microG phone, but still countless purpose-tailored devices that perform some proprietary function. If those proprietary functions (eg Google Voice SIP integration) could be straightforwardly done with libre software, then I could just consolidate those functions into the libre infrastructure.

> Taking a device that is supposedly mine and configuring it to work against my interests repudiates the idea that I am the owner of it

strongly agree here!

Save the planet for the children!

That link doesn't even challenge the claim, it just fails to completely verify it.