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

Ask HN: Risk of unsafe software in automobiles?

news.ycombinator.com

Post reply on HN
Print view View original

71–80 of 133 posts

Re: Ask HN: Risk of unsafe software in automobiles?

#71

Post by smt88 »

manscrober wrote: post #46
smt88 wrote: post #32

Don't buy a Tesla and you'll be fine. I work in auto insurance and the other OEMs actually care about safety and testing of software. Tesla has the most bugs by far.

I don't know about tesla since I haven't had a chance to drive one yet, but the number of bugs especially in the interface but also in driver's "assistance" in other cars(VW, BMW) makes me doubt that it makes a noticeable difference - at least with tesla there is a chance they will fix it

Other OEMs fix bugs all the time.

The difference is that Tesla has bugs throughout its software, including critical systems.

The other OEMs are extremely slow and methodical about updating things like their fuel-injection software, and that stuff goes through incredible amounts of QA. The same is not true of Tesla. They're a company that, at Musk's direction, do not prioritize QA or human life.

Re: Ask HN: Risk of unsafe software in automobiles?

#72

Post by simonbarker87 »

simonbarker87 wrote: post #70

Thoughts from my wife who has worked in electrical and software for OEM automakers (high volume, luxury sport and start up) for 10 years: (I’m typing while she is, ironically, driving our Volvo) To answer your last question first, buy a car that hasn’t been launched within the last 12 to 18 months. That’s not software specific, that general vehicle safety across the board as they will be working through the initial w…

Follow up to this:

The one thing that could cause a lot of problems for cars and software is Agile/Scrum.

The projects that are being run in this, new for the industry way, are always late and people hate working on them.

CEOs and other C suite people see the massively shorter lead times that software can offer and are getting greedy. They saved a year or more of time on a feature thanks to code and over the air and then they decide they want it made in 4 weeks, when 3 months would be prudent.

There’s something about the intangibility of software that makes traditional automotive people’s brains break.

Thankfully many rank and file engineers and PMs in OEMs are pushing back against Scrum etc so a more pragmatic layer of management will come up in the coming years. Sadly Agile/Scrum will cause some preventable issues in the meantime.

Unlikely to be safety critical stuff due to the rounds of QA and safety council sign offs and gateways they need to go through. But less safety critical stuff may slip through.

Re: Ask HN: Risk of unsafe software in automobiles?

#75

Post by ModernMech »

mikehodgson wrote: post #56
SoftTalker wrote: post #16

Earlier quoted context omitted.

My big complaint is with the transition of controls from dedicated, tactile knobs, switches, and levers to touch-screen buttons or menus which demand more visual attention (i.e. distraction from driving) to operate.

Mazda did the research, and transitioned back to real tactile controls in 2019. I expect more automakers will be or are already following their lead. https://www.motorauthority.com/news/1121372_why-mazda-is-pur...

I have a 2016 Mazda and they had tactile controls then. What they did in 2019 was remove the touch screen feature on the center display, but there were always tactile controls to access the display. I have a 2022 Mazda and it's largely the same interface as the 2016 controls.

Re: Ask HN: Risk of unsafe software in automobiles?

#77

Post by filoleg »

notdonspaulding wrote: post #17
izzydata wrote: post #8

All of that firmware that isn't a self driving autopilot blackbox is probably quite safe. Such as anti-lock breaks and fuel injection timing. I am never going to put my life in the hands of some software doing image analysis using machine learning.

> I am never going to put my life in the hands of some software doing image analysis using machine learning. Well...on your car, at least. I'm not sure how comforting that approach is when you're surrounded on the interstate by Tesla "FSD"s.

Ironically, I feel at much more ease driving surrounded by Tesla "FSD"s, as opposed to being surrounded by average Seattle drivers. At least FSD always uses turn signals when switching lanes and is way more cautious (while still remaining reasonable).

Back when I spent a full year commuting to work daily on a motorcycle, Teslas with "FSD" were the least of my worries. All while people who sharply switched into my lane in front of me with no turn signals used, they ended up almost killing me a couple of times.

Re: Ask HN: Risk of unsafe software in automobiles?

#78

Post by someweirdperson »

f1shy wrote: post #64

No amount of MISRA or ISO26262 or tests or any kind will help if the people doing the software are direct out-of-the-university mechanical engineers or physicist, who had at most one or two semesters of some kind of programming course. The reality is that this is the current state of affairs. Most of people doing software for cars have not the foggiest idea what software is really about. All the software I read is ju…

> Some examples I've seen in code

You are not reading it correctly. It is not code as everyone knows it. It's like an electrical circuit with variable names attached to each conductor, and the code propagates information like electricity would.

There's tools dedicated to this, able to draw pictures of such code circuits (e.g. Simulink, Ascet). And such pictures can be automatically translated into c-code, that looks even worse than anything translated manually.

In the end, of course the tests prove that the code works like the picture of the circuit shows, and therefore the car must work correctly! This avoids the need for anyone working on only the code to understand a car.

In reality, things usually work in the end only because of how simple everything is and high number of iterations.

Re: Ask HN: Risk of unsafe software in automobiles?

#79

Post by userbinator »

vel0city wrote: post #7

Your link of Sudden unintended acceleration contains a lot of entries related to this issue which don't involve computers at all. It lists pedal misapplication, entrapped pedals, stuck throttles, electrical shorts, and diesel engine runaway as other things which can cause such an issue. A lot of the reported incidents had nothing to do with software. Either way, if you've had a fuel injected car you were still expose…

Either way, if you've had a fuel injected car you were still exposed to these issues. You would have to go buy a carbureted engine from the 80s or before to get away from these "unintended acceleration" issues, as in the end a car with EFI probably has a computer actually controlling the injection.

Even with EFI, if the throttle is mechanical and the EFI continues to ask for more fuel for whatever reason (or a fuel injector gets stuck open), all that will happen is the engine will stall due to the excessively rich mixture.

Ever have vacuum hoses fail on an old car? Carburetors get stuck or clogged?

The normal failure mode of a carburetor leads to an engine that doesn't run, and not the opposite. Before complete failure, you will notice a performance decline.

Personally, I prefer no computer control.

On top of that I'll also get much better efficiency and reduce harmful emissions which hurt my family and my neighbors

You can get a lot better efficiency from a carbureted engine than most people think.

As for safety, I'd rather have freedom.

Re: Ask HN: Risk of unsafe software in automobiles?

#80

Post by userbinator »

simonbarker87 wrote: post #70

Thoughts from my wife who has worked in electrical and software for OEM automakers (high volume, luxury sport and start up) for 10 years: (I’m typing while she is, ironically, driving our Volvo) To answer your last question first, buy a car that hasn’t been launched within the last 12 to 18 months. That’s not software specific, that general vehicle safety across the board as they will be working through the initial w…

Nothing is perfect but the safety downsides of an old car are widely considered to be far greater than the threat of hacking or bad code in a new car.

"widely considered" by the same industry who would love to sell you a new car...

Post reply on HN