gpcz

Good luck finding every contributor to the Linux kernel and getting them to agree to that.

You would probably get what you want with a Deming regression.

Feynman learned calculus from the textbook "Calculus for the Practical Man".

This is the Hacker Newsiest comment I've ever read on Hacker News.

In functional safety, probabilities are usually clamped to an hour of use.

For now. If someone makes a practical quantum computer, pretty much every asymmetric primitive we use at the start of a cryptographic protocol to make a shared secret for symmetric encryption will be breakable.

Looking for a nice, solid, well-documented library to do something is difficult for most stuff. There are some real gems out there, but usually you end up having to roll your own thing. And Lisp generally encourages rolling your own thing.

I'm interested in knowing how they achieve requirements traceability to code and the requisite level of code coverage (certain SILs require MC/DC coverage, which usually requires expensive tooling). Also which hazards they identified and their mitigations.

Even if true, it seems like they're doing a pretty good job on their own.

The remaining 11 should see this as a sign of what's to come and quit.

Did it know where it isn't?

I believe that software-related college degrees are mainly there to get the horrible first few tens of thousands of lines of code out of people before they go into industry.

This reminds me of Joe Kennedy (JFK's father) getting out of the 1920s stock market right before the Depression because his shoe-shine boy was giving him stock tips.

"The same thing will happen if you're running a startup, of course. If you do everything the way the average startup does it, you should expect average performance. The problem here is, average performance means that you'll go out of business. The survival rate for startups is way less than fifty percent. So if you're running a startup, you had better be doing something odd. If not, you're in trouble."

http://www.paulgraham.com/avg.html

I have a lot of my old code, but I never want to look at it because it's super-cringe.

I host a simple tutorial on the Web, and the emails I got from people were...interesting. Some people assigned properties to the filter that were best described as magic. For example, trying to compute a theoretical true interest rate, trying to predict the output of a wind farm a year from now, etc.

It was hard to convince them that it's doing a pretty simple thing that won't do the things they wanted.

There may be a security advantage to using a separate non-bypassable network appliance that puts your traffic on Tor, since then it would be much harder to break into a Tails machine and make it leak your location. However, given that it's meant to be easy to use, I think they probably picked the right balance by having the Tor redirecting occur in the same address space as the computing environment.

Do they want Gerrymandering Simulator 2024, Bribe-A-Senator, or Citizen's United: SuperPAC Simulator?

I'd assume the Toyota Production System and their method of lean manufacturing.

Aviation safety for FAA airworthiness generally follows SAE ARP4761, which requires risks of Catastrophic severity to have a quantitative failure rate of 10^-9 incidents per hour or less. The Air Force follows MIL-STD 882E, which generally requires risks of Catastrophic severity to have a quantitative failure rate of 10^-6 incidents per hour or less. In short, the military accepts more risk than commercial aviation. That being said, military aircraft usually are also rated to ARP4761 levels of safety so they can fly in US airspace.

Hair dye that you purchase and apply yourself.

Instead, the companies want people willing to lie on forms.

All death is "fractional" if you make a fraction out of it. For example, dividing the number of people that died making the Empire State Building by the total number of workers. It seems subjective which fractions matter and which don't. It seems to be based on how much the responsibility can be laundered.

SeL4 uses capability-based security, but unfortunately it's not in common use.

From an ISO 26262 perspective, it seems like for the driving scenario of "unexpected windshield wiper speed increase", they may have set the controllability to C0 (Controllable in general) [1] when it may have needed to have been something less controllable (C1 means 99% of drivers can avoid harm in the condition, C2 means 95%, C3 means less than 90%). That probably set their risk level to something lower than it actually was. They probably need to do some tests for anything they're diverting to the touchscreen to figure out how many drivers can manage the situation within the time frame of avoiding a crash.

It looks like Tesla's job postings reference ISO 26262 experience as a plus [2], but I can't get a straight answer to Tesla's adherence to it in-house. My understanding is that no government mandates its use, but lots of car companies adopt it anyway.

[1] ISO 26262-3:2011(E) Table B.4 [2] https://www.tesla.com/careers/job/software-integrationengine...

Most safety standards that address software's contribution to hazards use some notion of "controllability" to determine risk. For example, ISO 26262 explicitly calls it controllability [1], and it directly contributes to how rigorously the software functionality has to be tested.

MIL-STD 882E uses "Software control categories" that determine the risk introduced by software [2, page 15-17]. Since those definitions are hard to track, they also provide a table near the end that tries to be a little more specific [2, page 96].

In other words, the safety community already takes into account the need for human or mechanical intervention when it comes to risks introduced by software.

[1] https://neweagle.net/how-iso-26262-2018-update-affects-you/ [2] https://www.dau.edu/cop/armyesoh/DAU%20Sponsored%20Documents...

Previous Discussions:

- Nov 26, 2010: https://news.ycombinator.com/item?id=1942778

- Aug 13, 2012: https://news.ycombinator.com/item?id=4374865

- Jul 5, 2016: https://news.ycombinator.com/item?id=12034277

- Nov 10, 2018: https://news.ycombinator.com/item?id=18421877

In "The Soul of a New Machine," there is an engineer who spends months debugging nanosecond-level glitches in their new CPU, snaps, and runs away after leaving a note: "I am going to a commune in Vermont and will deal with no unit of time shorter than a season."

The alleged Socrates quote is a mutation of some quotes in a dissertation published in 1907 where the author was summarizing common complaints about children in ancient times (src: https://quoteinvestigator.com/2010/05/01/misbehave/ ).