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

Autoland saves King Air, everyone reported safe

avbrief.com

Post reply on HN
Print view View original

121–130 of 193 posts

Re: Autoland saves King Air, everyone reported safe

#121

Post by vjvjvjvjghv »

jacquesm wrote: post #102
vjvjvjvjghv wrote: post #77

Earlier quoted context omitted.

I could be fast and impactful. Just in a negative way. The problem is that I come from the software dev side so I tend to be less interested in the medical side. It’s the same in a lot of safety critical. There is a lot of mundane work to tick the necessary checkboxes. There isn’t much that is interesting from a technological side. Maybe the result is interesting but getting there takes a lot of extremely boring work…

Maybe you should change your line of work. If you're that unhappy about what you do in spite of the fact that what you do is orders of magnitude more important than the next move-fast-and-break-things-advertising-driven-unicorn then that suggests to me that you should let someone else take over who does derive happiness from it and you get yours from a faster paced environment. Personally, you couldn't pay me enough…

I am retiring next year. So that should solve my problem :). I don’t know how other medical device companies are working but in mine leadership is dominated by people who know medical devices from a sales or medical perspective. Software is kind of secondary to them although it’s becoming really important. A lot of our processes aren’t very good for software so we end up doing a lot of work that makes no sense and makes the product actually worse. It’s better not to fix bugs because a new release will take months of paperwork. The requirement structure doesn’t map to software but the SOP isn’t written by people who known software. It feels a little like the development speed of NASA with the SLS vs SpaceX who are basically doing everything faster and cheaper while still having high reliability . My company is NASA here. Just very frustrating

Re: Autoland saves King Air, everyone reported safe

#122

Post by vjvjvjvjghv »

SoftTalker wrote: post #90
sinuhe69 wrote: post #60

Earlier quoted context omitted.

What is in this particular case that requires outdated tools? If they are code, certainly you can write them on VS Code or whatever you likes, and only need to compile and load on the original tools, can’t you?

You need tracability from requirements down to lines of code. It's a very painstaking process.

Painstaking and often done with terrible tools and badly written requirements.

Re: Autoland saves King Air, everyone reported safe

#123

Post by shrikant »

MuffinFlavored wrote: post #76
kylehotchkiss wrote: post #12

If you're one of the many developers at Garmin who worked on this, I can't imagine a better Christmas gift!

[flagged]

"This person sent goodwill to one group of people so obviously wants other groups to die in a fire" is not really devil's advocate...

Re: Autoland saves King Air, everyone reported safe

#124

Post by pjmlp »

gpm wrote: post #95
Palomides wrote: post #64

Earlier quoted context omitted.

it kinda sucks, but with the constraints it's at least understandable. they wanted an extremely lightweight language with a bytecode VM which could be ported to whatever MCUs in 2015, while also strictly limiting the functionality for battery usage reasons (and, uh, product segmentation/limiting third party access).

These days I'd say "sounds like wasm" but I guess 2015 was a bit before that took off.

There have been enough bytecodes since UNCOL in 1958 to chose from, and embedded is full of them, nothing special about WASM.

Re: Autoland saves King Air, everyone reported safe

#125

Post by pjmlp »

ilikehurdles wrote: post #51
ultrarunner wrote: post #42

Earlier quoted context omitted.

Seems like Java is popular at Garmin.

While I might not trust C code more than Java in life saving equipment, I would trust a median C developer over a Java one.

Given the amount of CVEs that would be a bad bet.

High integrity computing is full of pain staking processes, exactly because no one trusts C developers to do the right thing.

Re: Autoland saves King Air, everyone reported safe

#127

Post by AlotOfReading »

OptionOfT wrote: post #101
vjvjvjvjghv wrote: post #36

Earlier quoted context omitted.

“ Imagine getting to go to work everyday to work on something that actually saves lives.” I work on medical devices that improve and save lives but the work actually kind of sucks. You spend most of your time on documentation and develop with outdated tools. It’s important work but I would much prefer “move fast and break things”. So much more interesting.

Well, I'm glad it's that slow. I can't shake the idea of the horrors it would be to get a glucose pump whose software has been vibe-coded.

I work on team managing safety critical code. Management has asked to increase our AI usage, especially for generating requirements.

Re: Autoland saves King Air, everyone reported safe

#128

Post by wkat4242 »

shrubble wrote: post #112
Animats wrote: post #96

This is Garman SafeReturn, and this is its first real save. Here's a demo.[1] It's been shipping since about 2020, originally on the Cirrus Vision Jet. There's a lot going on. The system is aware of terrain, weather, and fuel, but not of runway status. So it gives the ground a few minutes to get ready, sending voice emergency messages to ATC. If you watch the flight track, you can see the aircraft circle several time…

Famously the golfer Payne Stewart and the total of 6 people on the LearJet 35, died after a sudden loss of cabin pressure incapacitated everyone including the pilots. A system like this, would have detected it and possibly saved them. I wouldn't expect a whole lot more detail, as that airport is often used by defense contractors like Ball Aerospace, who have a large office nearby.

SafeReturn doesn't detect that as I understand it. It still requires manual activation by one of the passengers.

Re: Autoland saves King Air, everyone reported safe

#129

Post by jacquesm »

vjvjvjvjghv wrote: post #121
jacquesm wrote: post #102

Earlier quoted context omitted.

Maybe you should change your line of work. If you're that unhappy about what you do in spite of the fact that what you do is orders of magnitude more important than the next move-fast-and-break-things-advertising-driven-unicorn then that suggests to me that you should let someone else take over who does derive happiness from it and you get yours from a faster paced environment. Personally, you couldn't pay me enough…

I am retiring next year. So that should solve my problem :). I don’t know how other medical device companies are working but in mine leadership is dominated by people who know medical devices from a sales or medical perspective. Software is kind of secondary to them although it’s becoming really important. A lot of our processes aren’t very good for software so we end up doing a lot of work that makes no sense and ma…

I've worked with a startup in the medical device space. Well funded. They were indistinguishable from most other startups, except in one detail: they did everything right. They made some extremely high tech stuff, very lightweight, and technology wise they were closer to watchmakers than to software and hardware people. I loved working with them and helped them to improve their yield (their QA was so strict that of their initial couple of runs more than 2/3rds of the devices got binned for the smallest infractions).

I suspect you may have just been unlucky with where you ended up. I'm getting closer to retirement myself but I no longer have to work for 'the man' so in that sense I got really lucky. But I really sympathize with how you feel. So, count the days, and look forward to something nicer. Best!

Re: Autoland saves King Air, everyone reported safe

#130

Post by silisili »

Animats wrote: post #96

This is Garman SafeReturn, and this is its first real save. Here's a demo.[1] It's been shipping since about 2020, originally on the Cirrus Vision Jet. There's a lot going on. The system is aware of terrain, weather, and fuel, but not of runway status. So it gives the ground a few minutes to get ready, sending voice emergency messages to ATC. If you watch the flight track, you can see the aircraft circle several time…

This is fascinating.

My uncle was a pilot, and I asked him 15 years or so ago about the job. He was going on and on about computers and autopilot, claiming that pilots were only really needed anymore for takeoffs and landings, and they could sleep during the rest. Probably realizing the liability in what he said, he was quick to clarify that he didn't, of course.

In that short time span we now have a system that can land a plane by itself. Nothing less than magic, and huge congratulations and thanks to everyone at Garmin who made this happen.

Post reply on HN