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

Internet-connected cars fail privacy and security tests conducted by Mozilla

gizmodo.com

Post reply on HN
Print view View original

431–440 of 660 posts

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#431

Post by lelandfe »

eximius wrote: post #383

I don't understand. This seems clearly illegal to have policies like this that claim consent by sitting in the car but the "user" has never been presented the policy in a way to consent to! Is this just a good lawsuit away from being thrown out but no one has done it? Is there some particularly fucked up legal precedent that makes this tenuously legally stable?

> You promise to educate and inform all users and occupants of your Vehicle about the Services and System features and limitations, the terms of the Agreement, including terms concerning data collection and use and privacy, and the Nissan Privacy Policy

Similarly, how is this kosher? Do we think anyone, even a single person, has ever followed through on this promise they’ve apparently agreed to in using a Nissan?

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#432

Post by cduzz »

sergiosgc wrote: post #391
ImprovedSilence wrote: post #357

Earlier quoted context omitted.

drivers that enjoy manuals have always been around. We’re just louder now because they almost don't make cars with manuals anymore, so when one does come along we rejoice.

Genuine question: What exactly do you enjoy? Twenty years go, you could do a mountain road engine braking with downshifts, getting then the perfect gear for the turn apex and coming out perfectly balanced. But cars have changed. Engine braking is a lot less effective today (different compression ratios, better mechanics). Automatics now have more gears and allow you to manually select the gear, so you can control ava…

I dislike indeterminate lag between input and action.

With an automatic, there's a threshold where the car decides to downshift when asking for a particular increase in forward velocity; that set point will wander depending on current RPM state and velocity and drive gear ratio.

Modern cars are bad enough with turbos and fancy valve timing and throttle by wire stuff where the behavior of the thing is a big stack of jitter, but adding a transmission to the mix makes the response times even more random.

At least with a manual transmission, the behavior of the throttle pedal is far more predictable and direct -- down the engine will go faster (modulo the current drive gear) and up the engine will slow down and slow the car down. Often you're in the incorrect gear for a particular desired acceleration but there's a feedback loop that you participate in to recognize / avoid the issue (mash pedal, not much happens because you're in the wrong gear, you get feedback and decide to change gears). With an automatic, you're just yelling down to the engine room asking the hamster to get on a different wheel.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#433

Post by hedora »

toastal wrote: post #97
vannucci wrote: post #78

The more I read about these things, the more I think I'll be driving my 23 year old Toyota 4Runner until the end of my life

I switched to a motorbike that has almost nothing ‘fancy’ about it--along with public transportation.

If you have a license plate, then its being tracked with cameras when you use it.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#434

Post by bigbacaloa »

yafbum wrote: post #231

The article is not very clear about whether this kind of abusive data collection is actually happening and they can prove it, or whether they found the text of the privacy policies to be overly broad. I have no idea how a car would infer my sexual orientation.

That's easy. But they know that already anyway.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#435

Post by lotsofpulp »

EricE wrote: post #387
TheRealDunkirk wrote: post #223

A lot of people are making comments about defeating the telemetry, but it's not just about selling ads this time. This is about actual surveillance and (eventually) control, and the governments of the world will make this a requirement, and make circumvention of it as illegal as copyright infringement. In the US, the government will throw up it's hands and say it's not us; it's the "free" market, and this must be wha…

just wait until insurance companies demand access to your telemetry or they triple your rates.

Why would they triple your rates? Insurance regulators do not let insurers increase premiums unless they have evidence showing the new data indicates a higher risk factor.

So if you are getting your rates tripled, then you were being subsidized by safer drivers before, which seems like something that should be fixed.

Dash cams have already probably given insurers a better idea on who is a riskier driver and who is not since collisions can now be more accurately attributed to the at fault party.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#436

Post by hedora »

Nifty3929 wrote: post #144

The biggest problem I have is tracking that the government has access to. We're in a situation where the government knows everywhere your phone has been, and everywhere your car has been. I doubt we're far from the gov't being able to de-activate your car at will also. I would like to have privacy from corporations, but I really wish we'd keep the focus on privacy from our governments, since they have such ultimate p…

In the US, the government has access to 100% of the information the corporations have.

Also, corporations have all sorts of power over us that should be reserved for the government. For instance, they control electronic currency, freedom of association (online), and your ability to distributed and purchase most media.

The government has been able to use onstar to deactivate GMC vehicles at will for over a decade. Of course, GM can do this too.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#437

Post by ajross »

0cf8612b2e1e wrote: post #206
gonehome wrote: post #182

Earlier quoted context omitted.

[flagged]

I was riding in a friend’s Tesla where the screen crashed while we were on the road. Just a black screen for some amount of time and had zero insight into the state of the car. That just does not happen with physical controls.

> That just does not happen with physical controls.

I was driving a car once where the handle to the stick shift literally popped off in my hand. So... no, that's just silly. Stuff breaks. Important stuff breaks. You deal with that with careful design and redundancy[1], not whining on the internet about touchscreens.

[1] Like how in the Tesla all the driving controls are, in fact, NOT connected to the touchscreen controlled by the MCU but to the AP computer.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#438

Post by hedora »

0xcafefood wrote: post #13

It almost feels silly to ask, but is this legal even in the United States with its comparatively weak privacy laws? In many states, a vehicle is legally an extension of the home. So legal rights and protections that apply in one's home also apply in one's vehicle. Is the idea that, buried somewhere in the legalese, is a statement that the buyer is granting the automaker the right to spy?

The supreme court recently decided there's no right to privacy in the US (they used this as justification for rolling back Roe v. Wade). It's pretty clear they'll be taking away more rights over the next few years.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#439

Post by lotsofpulp »

renegat0x0 wrote: post #194

In modern era everything mechanical, or analog is being replaced with digital, or with something "as a software". Product is replaced with "service". When everything becomes in the end connected with the Internet, everything is dependent... A system crash of AWS, Google, or any other monolithic provider will have drastic results. A system crash could halt many cars in a country. In the future Hackers could stall enti…

> cameras in cars. What they will detect? If I am drunk, or something more?

The more would be if you are watching your phone, in 99% of cases.

Re: Internet-connected cars fail privacy and security tests conducted by Mozilla

#440

Post by HumblyTossed »

gonehome wrote: post #307
HumblyTossed wrote: post #199

Earlier quoted context omitted.

> It reminds me of the complaining in the early iPhone days about lack of a physical keyboard. This is so not the same thing.

It really is imo. Good touchscreen UI is superior in many ways (though not all) to fixed control buttons. It’s why it dominates phones and why it’s winning in the market for cars as well (Model Y was top selling car on earth for Q1 2023). Bad touchscreen UIs suck, but that’s also true of phones. Good tactile controls have some advantages, but ultimately they’re minor and worse on net than a good touchscreen interface…

>It really is imo.

It really is not IMO.

A touchscreen UI is a visual medium. That is not what you want/need while operating a vehicle. You should be able to change the AC by feeling for the button and not taking your eyes off the road.

This isn't just HN being HN again. Driving enthusiasts are pushing back on touchscreens.

Post reply on HN