Comments by "Traveller" (@traveller23e) on "Mentour Pilot" channel.

As a programmer, I know that no matter how hard you might try, it is in practice impossible to plan the optimal outcome for all possible scenarios in a system with any level of complexity. At some point the human in the system will act in a way you didn't expect, or the computer will kill one of the processes, or a server developed by a different team will return a web response you didn't expect. It's not always a matter of assuming something will never happen, it's just as often or more often that something just never occurs to anyone as a possibility. And yes, there are plenty of software houses that rush stuff to market, there's plenty of code that assumes a computer's clock time won't be set to before the Unix epoch and just figure if it's an issue the code can be patched, but when it comes to systems with a high importance such as these, that require a lot of certification and so on, it's probably best to assume the mistakes weren't made intentionally to cut costs unless such a motivation and deliberate decision can be proven.
10
@biggiouschinnus7489 Surely the evolution of planes though wouldn't effect a pilot's instinct to cross-check all sources of the same data though...
3
@DowncastParadox Yeah, definitely and that was something mentioned in the final report as well. I just take issue with the "critical to perfectly anticipate everything, as I know from my decades of experience" thing, as if anyone can perfectly plan anything above trivial complexity. Maybe I'm overly sensitive to it, having once had a boss with just that thinking who'd take any opportunity to belittle his employees and brag about how much faster than us he would be able to do things😅
2
@nomore6167 There's a difference between logging "Value should_continue was unexpectedly ja, returning -1" and "User typed ja instead of the expected y or yes, but we'll assume that means yes anyway". What I'm saying is that your implication that the people responsible had made the deliberate decision of not handling that edge case is probably inaccurate. The edge case was definitely handled somehow (i.e. goaround thrust was not applied and there weren't any obvious warnings to the pilots), however as the report found in this particular case the somehow wasn't perfect. Likewise, had the humans in the system (pilots) been trained for this exact scenario and acted according to that training, the issue would have been averted. As I previously pointed out, unless there's actually evidence that people noticed the potential shortcomings in the system (in both the human and computer portions) and decided it wasn't worth investing effort into, it's probably safer to assume that it just never occurred to anyone as a possibility.
1
@johnsteward8325 I'm not in the US, but I can say that the market's pretty good more or less everywhere for programmers. Especially the fact you're full-stack is great- everyone's looking for those. As for not having a bachelor's, I don't have one either and I haven't had too much of an issue. Again, I'm not in the US and actually currently I consult for a company that can't hire me directly due to an ancient policy against hiring people with no degrees. It's mostly fine, though. I would definitely lean on your three years of experience though, not your schooling. They're even advertising "senior" roles with very few years of experience. If you're able to describe your projects in detail or show them off at all, that might help too, but perhaps best of all would be to learn to market yourself. Chances are the first round interviewers won't know a thing about programming, so if you're downplaying what you've done they probably won't even pick up on it. Also, watch out for any red flags...there are a lot of terrible companies out there so do what you can to avoid them. Sorry I can't be more helpful, but I guess the good news is you should be able to find something.
1
@johnsteward8325 Ah, a couple technology-related things I forgot: if you're knowledgeable in Java, changing to C# would be super easy so don't discount C# roles as well. Peer review is a good thing, ask companies what their process is and make sure they have one in place. And take it with a grain of salt if someone tells you to always or never do something, or goes into programming philosophy. Chances are the pattern you're implementing isn't what was originally intended by whoever was a proponent of it back in the day, and even if it is it might not suit your case in the textbook form. As for always and never, they often lead to all kinds of grey areas the person giving you that instruction didn't consider or potentially know was a thing. When it's regarding a language feature (classic example: goto) it's often the result of something being considered bad form back in the day, and that being blown out of proportion until long after the original issue with the feature or pattern was largely forgotten and didn't exist anymore.
1
Honestly having watched all of his episodes on accidents and incidents, this is the one that's scared me the most. The fact there's such a big fire risk right at the only part of the plane from which it can be controlled...it's amazing this is the first time there's been major accident because of it.
1