Earlier quoted context omitted.
OK? I don’t think that sounds like an HN commenter at all.
The other thread delivers. https://news.ycombinator.com/item?id=37406265
Internet-connected cars fail privacy and security tests conducted by Mozilla
641–650 of 660 posts
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#642This is very much in line with the industry trying desperately to get into the tech recurring revenue model. > the manufacturer collects information including sexual activity, health diagnosis data, and genetic data, though there’s no details about how exactly that data is gathered How is that even possible? Without more details about what this means and how it's done, this sounds a bit fishy to me. > Mozilla said it…
> How is that even possible? Yesterday, you visited a dating site a bunch. Today, you drove to a house that wasn’t yours and the passenger side door opened. Then you drove to a restaurant and the passenger side door opened again. Then you drove back to the house you don’t live at and the car was there overnight.
https://www.reuters.com/technology/tesla-workers-shared-sens...
they have interior cameras too, above the rearview mirror.
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#643Earlier quoted context omitted.
The other thread delivers. https://news.ycombinator.com/item?id=37406265
Keep on donating to Mozilla so they can keep doing these little pet projects with absolutely no impact while paying executives millions of dollars.
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#644Earlier quoted context omitted.
A 2022 Honda Passport has physical buttons for pretty much everything. The only thing I use the touchscreen for is as a display for maps from CarPlay (and the backup camera). So reliance on the touchscreen for things like environmental controls isn't universal even in new cars.
I love Honda interiors. The base models are almost always the best with big unique knobs and buttons. However, I'm really disappointed on their electrification. I would speculate that their misguided attempt to push Hydrogen is going to kill the Japanese companies if they don't invest in BEVs soon. I would seriously consider a 300 mile range, electrified Honda Fit using Chevy's Ultium platform. But it seems car compa…
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#645Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#646Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#647Earlier quoted context omitted.
Yup. Here's a fun hack - you can drive car in EU on US plates, due to international agreements. In that case, you don't have to follow local car inspection standards, but inspection standards of your home country. Get a plate from US state that has no inspections? You need no inspections at all!
There's usually a time limit on that. Last I checked in my country it's six months before you have to register the car, get local plates, and pay VRT (vehicle registration tax) + VAT. Including the cost of VRT in the VAT calculation, resulting in double taxation. You can't put foreign plates on an irish registered car without first bringing it out of the country and registering it abroad either. I don't think there's…
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#648Earlier quoted context omitted.
What? Apparently car accidents aren’t a thing near you. “Nine times out of ten” is a wholly made up and entirely inaccurate estimation. “Sure, someone just totaled my car but I can “just come back later when the market changes”.”
If you think my numbers are made up, take the effort to refute them. Are you saying 9/10 people who buy a car are buying it because they got in an accident? There were 13.7 million new cars sold in 2022. I doubt you'll find 13.7 totaled cars in 2022. And it still stands to reason that if somebody got in an accident, why would they turn around and buy a new car with a big mark up and massive waiting period when they c…
So I'm required to make effort to refute numbers, but you're not required to make effort to substantiate them? Huh...
> Are you saying 9/10 people who buy a car are buying it because they got in an accident?
I never said any such thing. I just think the number is higher than 1/10.
> There were 13.7 million new cars sold in 2022. I doubt you'll find 13.7 totaled cars in 2022.
I wouldn't expect to. But I'd definitely expect the number to be over 1.37 million.
Hurricane Ian alone resulted in 400,000 vehicles damaged[1], and that was just one environmental event, before we look at collisions.
Washington State (with 2.3% of the US population) reported 104,000 collisions [2]. So we might extrapolate to somewhere in the order of 5 million collisions a year. Indeed, that number is probably closer to 6 million [3].
So we're already approaching 7 million vehicle 'incidents' a year including collisions and one 'act of god'. I think it's reasonable to set a number of 8 million when you factor in all the other tornados, hurricanes, hail storms and others.
It's hard to determine what fraction of those are total losses. However...
US insurers paid out $173B in losses. [4]
$173B across 8 million incidents is around $21,000. Now in a good number of those losses there are 2, or more vehicles involved, which increases the number some. But even if we assume that say 50% of collisions involved another vehicle, then we have an an average auto loss of $10K per vehicle per collision.
Most insurers work on a 70/75 rule for total loss, that is, if the cost to repair the vehicle exceeds 70 or 75% of its value, it is written off.
So if your collision results in "only" a $2-3K repair bill (which can be cheap these days), then you can see that there's a notable amount of payouts on total losses. (It's also hard to get a value of the average vehicle on the road in a given year - insurers are the only ones who'd likely be able to supply that information, and it would be buried somewhere in actuarial tables more than anything).
It's not 9/10. Nor did I ever say it was.
But if I had to make an estimate, I'd say it was closer to 2 or 3/10.
[1] https://www.automotive-fleet.com/10183423/carfax-estimates-h...
[2] https://www.weierlaw.com/2022-washington-state-car-accident-...
[3] https://www.simplyinsurance.com/how-many-people-die-in-car-a...
[4] https://policyadvice.net/insurance/insights/auto-insurance-s...
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#649Earlier quoted context omitted.
American carbrains can’t imagine a society that doesn’t depend on huge ass vehicles for daily transportation. A reminder that driving isn’t a right, it’s a privilege that you have to get a license to do, and many other places that aren’t America don’t design their cities and even their small towns [1] around the idea that you must own a vehicle. Congestion taxes and pollution rules tend to affect city centers where p…
"A reminder that driving isn’t a right" It is in America. Our constitution constrains the government, it doesn't grant us rights - we already have them as human beings. I can't imagine living under a cynical government that has that equation flipped but obviously you've been conditioned to accept it.
Re: Internet-connected cars fail privacy and security tests conducted by Mozilla
#650Earlier quoted context omitted.
20 years from now there will be companies making dumbcars just like there are starting to with phones right now
The problem is complying with legislation. This is worse than phones.
I think some already comes with that.