HNHacker News
TopNewBestAskShowJobs

scarier

576 karma · joined July 2, 2020

submissionscomments
scarier··on The toll of dating app burnout
I’m not convinced it’s a pure numbers game, at least from a perspective of long-term success (for what it’s worth, I was averaging 1-2 dates/year until I met my partner). I think it’s much better to have two high-quality dates than dozens of low-quality ones. The advantage of dating apps over more traditional ways of meeting people is that you can much more effectively set up bandpass filters to focus your efforts on interesting people.

I think the best thing you can do is focus on yourself, particularly by cultivating hobbies that you enjoy—bonus points if they’re things you can enjoy alone and with others, because they’re easy inroads and great first dates (worst case you get to spend a bit of time doing something you like!).

The second thing I’d encourage is to get any potential dealbreakers out of the way early (e.g. religion, political views, desires for parenthood, career aspirations). It’s a slow path, but it’s helped me avoid the major pitfalls I’ve seen in my friend’s’ and colleagues’ relationships.

scarier··on The Church of Interruption (2018)
Why are you so sure that free will exists?
scarier··on George Orwell’s 1940 Review of Mein Kampf
Experiences in war vary greatly, and feelings about them are complicated and conflicted. I've met many people who genuinely seemed to enjoy it. This article nails it: https://www.esquire.com/news-politics/news/a28718/why-men-lo...
scarier··on Lilium achieves first main wing transition for all-electric aircraft [video]
The small distributed motors on the X-57 are only used on takeoff and landing to augment lift. The small propellers fold away in cruise, when the entirety of the aircraft’s thrust is provided by two large propellers at the wingtips.
scarier··on Crabs evolved at least five times from separate groups of crustaceans (2021)
Relevant XKCD: https://xkcd.com/2314/
scarier··on Cargo sous terrain – Underground cargo train in Switzerland
CST isn’t built to solve the last mile problem either—it explicitly shows cargo being delivered to and picked up from central hubs.
scarier··on Cargo sous terrain – Underground cargo train in Switzerland
Because it needs far more complicated and expensive cars for no real benefit.
scarier··on Cargo sous terrain – Underground cargo train in Switzerland
The issue isn’t how automated it is— it’s easy to automate trains and automatically load pallets. Making each car responsible for autonomous navigation in two dimensions and self-loading becomes a much more complicated engineering problem.
scarier··on Cargo sous terrain – Underground cargo train in Switzerland
I’m always amazed every time someone tried really hard to invent new technology that’s just a worse train. I love the overall concept, but damn they’re making it way more complicated than it needs to be.
scarier··on French fighter jet joy ride goes très, très wrong (2020)
I mean, it wouldn't be the first time someone forgot one or two of their FENCE checks...
scarier··on French fighter jet joy ride goes très, très wrong (2020)
I'm curious about that explanation--in the seats I'm familiar with, initiation (including all the different potential sequencing modes) is done entirely through an explosive train to prevent this exact failure mode (loose Cannon plugs cause plenty of other problems though).

We're talking about different things. In "forward both, aft self" mode a front-seat-initiated ejection still ejects both seats in the normal sequence, while in "solo" mode it would only eject itself. For obvious reasons, you would never fly with a passenger in "solo" mode.

scarier··on French fighter jet joy ride goes très, très wrong (2020)
This is supposedly where the idiom "bought the farm" comes from.
scarier··on French fighter jet joy ride goes très, très wrong (2020)
It’s wild that the ejection sequence failed. Some two-seat aircraft have a “forward both, aft self” ejection mode for these types of events where you don’t trust the guy in the back to make good ejection decisions, and I originally assumed that was what happened here. Oops.
scarier··on Residential building struck by a missile in southwestern Kyiv Saturday morning
That's a lot of damage against a reinforced concrete structure for an anti-air weapon. I think it's at least as likely to be an impact from an air-to-surface weapon, which would also have solid propellent. The flight profile looks much more consistent with a helicopter or fixed-wing aircraft launch than a surface-to-air launch (especially since, as you mentioned, it's clearly in the early-ish stages of flight). A Buk launch could be vaguely plausible (the Strela series has VERY small warheads), but I don't see the S-300 having that kind of failure mode early in flight.

As you said though, the full picture will come out later. Hopefully they'll find some identifiable weapon fragments eventually.

scarier··on My favorite liar (2009)
Frankly, even a working calculator can induce the same problem. My school allowed calculators with a “solve” function to be used on calculus tests, and a staggering number of people would directly transcribe whatever output it gave, regardless of how nonsensical. “NaN” was a favorite.
scarier··on My First Impressions of Web3
Was the Prius ever intended to be an evolutionary step toward widespread EV adoption?
scarier··on If – A Poem by Rudyard Kipling
Agreed. I'm well aware that he was a racist, jingoist prick, but "If-" is one of my favorite poems of all time. On the flip side, I have a hard time reading poems like "The White Man's Burden" and "Gunga Din" as anything but fairly clever satire. Maybe that's a measure of progress?
scarier··on Turboencabulator
Loosely related: a lecture on missile guidance laws...

https://www.youtube.com/watch?v=bZe5J8SVCYQ

scarier··on Turboencabulator
Somewhat related: https://www.youtube.com/watch?v=bZe5J8SVCYQ
scarier··on FAA investigating controversial crash video
If anyone's interested in watching videos of genuine emergencies in light aircraft, Elliot Seguin (https://www.youtube.com/c/utopiasnow) has some fantastic ones.
scarier··on Aviation Safety Whistleblower Report on 737 Max [pdf]
Heh, I think I did a poor job of expressing myself. You're absolutely right that an intermittent/erratic trim operation is a failure, and potentially a serious one. I'm thankful that you, as a flight control engineer, are as concerned about it as you are (for hopefully obvious reasons). Out of curiosity, how probabilistic is your failure mode analysis? I'm wondering what the relative likelihood of an intermittent uncommanded actuation compared to the neutral-to-extreme runaway failure that most pilots expect. I've never been in a sim where the operator console had trim failure options other than "stuck at current position" or "runaway to extreme limits," but I wouldn't be surprised if they should've included other failure modes.

Yeah, that seems like an eminently reasonable way to design a trim system. I don't think the MCAS concept is fundamentally unsound, but it blows my mind that they didn't design it with that kind of mindset.

Oh man, I wish more software was built to the standards of aerospace control system best practices...

scarier··on Aviation Safety Whistleblower Report on 737 Max [pdf]
In all seriousness, I would enjoy talking about commercial aircraft trim system failure modes over a beer sometime.

For what it's worth, while I agree with your technical definition of a trim runaway, every time I've seen it in the sim or real life it's been a single, continuous event moving from steady-state flight trim to an extreme. I'd be willing to bet a few beers that this is what most pilots are trained to expect from a trim runaway, and what B737 crew see in the sim while getting type rated. I'm not disagreeing that if the LA crew diagnosed it as a trim failure and performed the EP correctly they would likely still be alive, and I'm also not arguing that they were an exceptionally good or even average crew.

I'm arguing that you can't really fault a below-average-but-still-acceptably-competent crew for not diagnosing the failure of a system they couldn't have reasonably been aware of as a trim problem on an otherwise perfectly functioning aircraft. There are plenty of atypical emergencies that require the crew to "do some pilot shit" to get the plane back safely on deck, but an easily foreseeable single-sensor failure shouldn't be one.

We'll probably just have to agree to disagree about how likely an average crew would be to treat this as a trim failure, but I like to think we can still agree that the likelihood was unacceptably low for commercial aviation safety standards.

scarier··on Aviation Safety Whistleblower Report on 737 Max [pdf]
>Like this one distributed to ALL 737MAX crews:

...after their poorly designed system caused a plane crash.

If you have a 737 type rating and can speak to type-specific training standards I would love to be educated, but the defining characteristic of runaway trim failures that I've experienced is that the trim keeps going in one continuous motion until it can't go any further or you manually shut off the system. A momentary, uncommanded attitude change would initially make me want to troubleshoot the autopilot, rather than the trim system. This is exactly why you have to describe new systems and their failure modes in detail, even if they have elements in common with and use the same emergency procedures as existing failures.

scarier··on Aviation Safety Whistleblower Report on 737 Max [pdf]
I mean, that's all well and good, but assuming that your pilots will correctly diagnose and correct a failure of a new system that may or may not have similar symptoms to existing emergencies is still a really poor practice. If MCAS had been designed to command a continuous forward pitch moment until the AOA excursion had resolved, it would have very accurately resembled a pitch trim runaway following an AOA probe failure. As it was, it clearly didn't have a strong enough resemblance.

Safely operating a poorly designed aircraft can be done, but it starts with explaining the deficiency in exhaustive detail in a bold typeface in a prominent place in the operating manual, with clear warnings to avoid certain flight regions, and a well-documented emergency procedure. For example:

MCAS FAILURE: If you experience repeated momentary uncommanded nose-down pitch excursions. 1. Conduct RUNAWAY PITCH TRIM procedure. 2. Do not reset pitch trim circuit .

Unfortunately that would have required new training.

scarier··on Artificial Intelligence in the Cockpit
Thanks! Should be fixed now.
scarier··on Artificial Intelligence in the Cockpit
I think this lecture (https://youtu.be/5ESJH1NLMLs) should be mandatory for anyone trying to improve flight safety with more automation. I think we're already at diminishing returns, and the only way to eliminate accidents caused by over-reliance on automated systems is to cut the pilot out of the loop entirely. Even in general aviation, the low-hanging fruit is elsewhere.

That said, this is still pretty cool, and I could see something like it being one component of a much larger fully automated flight management system.

Edit: link should be fixed. If I'm still crushing it, the title of the lecture is "Children of the Magenta Line."

scarier··on Groups never admit failure
>Now, the US military is struggling with the "Fighter Mafia", which tends to run the USAF despite the ascendancy of drones and the usefulness of the A-10.

Duuuuude, the Fighter Mafia was active in like the '60s and '70s, and some of its members were involved in the design of the A-10 (in line with their strong advocacy against multirole aircraft). Do you have any evidence that current USAF strategic policy is dictated by the fighter community? Fighter pilots have lamented the loss of "l'audace" for decades, but despite their bravado I don't think any of them really believe that they deserve absolute strategic primacy.

scarier··on A submarine's weakness: its software?
Hypersonic glide weapon.
scarier··on A submarine's weakness: its software?
>Fractional Orbital Bombardment System is new and very scary for the USA missile defense systems.

It's actually a fairly old technology as these things go, developed in the '60s and deprecated in the '80s. My understanding of HGWs that are currently being developed/fielded is that their potential FOBS aspect isn't what's scary--it's that they're more maneuverable than MARVs on traditional ballistic missiles and may be able to evade many currently fielded ABM systems.

That said, submarine warfare is indeed insane.

scarier··on A submarine's weakness: its software?
The oldest aircraft navigational aids to avoid hitting the ground are fairly simple, and predate software (the radar altimeter is a good example, which in its most basic form just tells you your altitude above the ground and works really well in most flight regimes). Ground proximity warning systems are more along the lines of what you're thinking, and rely on the aircraft's flight computer to predict an intersecting flight path with its digital terrain model within a specified period of time (typically relies on GPS and/or INS data). As long as the software isn't integrated into aircraft flight controls (auto-GCAS in fly-by-wire military aircraft, for example) it doesn't necessarily need super-rigorous testing, although when it is connected to automated flight controls it should* have* (looking at you, Boeing...) an extremely rigorous testing process to ensure it behaves deterministically and as the pilot expects under all normal and emergency conditions. This generally involves a lot of bench testing, with a (potentially very large) number of test flights to ensure it works as designed in enough distinct conditions to have full confidence in the system.

TCAS (the family of systems aircraft use to avoid hitting other aircraft) is much simpler and doesn't rely as heavily on GPS/INS data, but generally only gives data on cooperative aircraft transmitting transponder or ADS-B data. A very simple version gives a traffic alert when another aircraft approaches within a specified distance of you, but may not give any other useful information (relative altitude/position, heading, etc.), and the most basic version I can think of just gives a range to a single cooperative aircraft.

I think you'll find that you're off by a few orders of magnitude on the number of submarine navigators out there, and there are few enough certified avionics systems that most are used by a substantial enough audience (and, more importantly, mandated by enough national aviation authorities) to be worth the R&D effort.

That said, military hardware is often decades behind the civilian world when it comes to many of these mandated safety features, so I don't find it surprising that military ships (often with proprietary radar and sonar systems designed primarily for combat, and not wanting to broadcast their location to potential adversaries) don't have good integration with commercial collision avoidance software designed to work in a collaborative environment with off-the-shelf hardware that may have only been installed as an afterthought, if at all.

← PreviousPage 3 of 6Next →