jcgrillo

Yes, that too

Admittedly I have nowhere near the flight hours, training, or expertise of these pilots, but having flown airplanes myself I can totally imagine in an off-nominal situation (which I have been in before) conscious focus is fully on flying the airplane even if your rote lizard brain is procedurally going through the motions of pulling the ejection handles or otherwise responding to the emergency. My instructor's words--he was a helicopter pilot (Hueys and Chinooks) in the Vietnam war with some 20k hrs logged in complex aircraft, jets, etc. since so I know for certain he knew wtf he was talking about--going through my head "do not ever stop flying the airplane". In this case, my conscious focus would be to stomp one of those rudder pedals as hard as I could to try to recover from the spin, even if I was also simultaneously yelling "eject" or whatever you're told to into the intercom and pulling the handles. But I haven't ever been trained to eject from an aircraft, or maybe my instinctive predilections would select me out of the training regimen these pilots go through.. who knows

Earlier today Siri notified me over and over again to message a particular contact on the GMail app. I have no earthly idea why this was important enough to notify me about, I can figure out whether I need to message people. It provided no hints what the contents of the message were supposed to be, or why I might need to message them now instead of some other time. 20yr ago when I worked in an industrial steel fabrication shop, a couple times a week someone would exclaim "shoot the fuckin' engineer that came up with that one!" usually regarding some bone-headed physically impossible weld on a plan or a procedure that would clearly result in an assembly being out of tolerance, but sometimes something more serious like a weak or dangerous design. Now we're well into the "shoot the fuckin' engineer" stage with tech products. Abusing people like this is wrong. Their attention is a finite resource. Their reward centers are vulnerable. Mindlessly pushing the buttons in the psychological control room of vast, diverse, and increasingly stressed populations is a profoundly stupid idea. If we're not careful somebody is actually going to start shooting the engineers. Would they be wrong to? I don't think the backlash against this shit is going to be small or subtle, and I'm honestly afraid to be associated with this industry right now. Y'all are playing with fire in the most reckless, clueless, thoughtless, and callous fashion. Be better. Stat.

If the Internet is to be believed they're not actually more expensive than an F/A-18, and as far as military aircraft go.. not the most expensive. But a ~$150M accident is nothing to sneeze at.

Arms and legs can take a serious beating too. Airplane cockpits are pretty tight spaces, and to be explosively shot out of one with little notice is.. yikes.

> acquire gently-used older models, keep them working, and run Linux.

Even as a financially secure mid-career engineer this is still an excellent formula. Buy a "retired" Thinkpad on Ebay, upgrade RAM and NVMe if needed, replace battery if needed, then run it for a decade or more.

I could have kept running the ca. 2015 T460p as my primary laptop--it has 32GB RAM, and decently powerful CPU and GPU--but the T14 is lighter, has much better battery life, generates less heat, etc. etc. Maybe I don't "get it" but I'm pretty happy with the fancy "new" machine.

I still have my x61s, T460p, and currently running a T14 Gen 3. The x61s is just sitting there awaiting some future project. Hopefully it happens someday. The T460p has become a homelab server. I never should have sold my W500, that thing was awesome. I also have an 11e which is my burner Windows machine for doing sketchy stuff with cars.

Yeah I think moving forward one of the questions I'll be asking companies I interview with is "what does your seniority distribution look like and how do you intend to maintain it?"

I'm sorry but both of these are false equivalences. CNC isn't about making general machining operations faster or necessarily better. It's about making a single machine more versatile. Instead of needing an assembly line of machines you can get a bunch of different operations done on the same part without moving it to a different machine. You can also do compound operations that were otherwise highly specialized (like milling a turbocharger's radial compressor wheel). You can get the same job done with a series of manual operations though.

A garden hose vs a bucket is also the same situation. You can accomplish the same thing with either, but one might be more labor intensive.

AI is nothing like either of those. It would be like instead of a bucket you get a garden hose that points in a different direction every time you try to use it. Or instead of a 5 axis mill that rigorously executes the g-code it just randomly reinterprets tool paths each time it cuts a part. Both of these things would be worse than useless in their respective applications.

AI is different because it plays to the pliability of the software domain. Even fairly shitty, irreproducible results can be good enough for software development, if you don't look at it too closely. Make analogies to the physical world at your peril!

Couldn't have said it better myself.

Nor is the LLM output tightly coupled to the training data. They'll "eagerly"[1] fill in the blanks wherever it sounds good.

[1] here I don't mean to imply agency, just vigor.

Crocodile tears, just like the fake "fear" of its capabilities. Anything to raise another round of dumb oil money.

Like this: https://en.wikipedia.org/wiki/Times_Beach%2C_Missouri

Scrape off all the soil, put it in casks, and bury it in a concrete bunker for 10000 years. Then relocate everyone and attempt to rebuild.

jfc lmao

> Somewhere in the future, the new software engineering will be primarily about principles to avoid this in the first...

It's really nowhere near as complicated as making distributed systems reliable. It's really quite simple: read a fucking book.

Well, actually read a lot of books. And write a lot of software. And read a lot of software. And do your goddamn job, engineer. Be honest about what you know, what you know you don't know, and what you urgently need to find out next.

There is no magic. Hard work is hard. If you don't like it get the fuck out of this profession and find a different one to ruin.

We all need to get a hell of a lot more hostile and unwelcoming towards these lazy assholes.

> because that’s not how LLMs work

It's almost impossible to have a rational discussion about the effects of this technology because this point is so easily lost. Even super smart, credentialed, expert people easily (and often!) fall into the trap of anthropomorphizing the bot because it makes human noises. It's really important to remember the mechanical principles underlying its function. No different from any other computer program in that respect, the difference is the psychological hold it gets on the user. There is no intention behind its actions, but it's very easy to hallucinate one because with every other thing that speaks human language there is some intention behind the words and actions.

They'll often exploit a power dynamic. If you're less senior than them, in some organizations it can be very difficult to stand up to that behavior. I experienced that at my previous employer as a relatively senior engineer, even. Top down organizations look unfavorably on feedback that runs upwards. Your pushback will be seen as standing in the way of progress.

Right, so the Github "open contributions" model where anyone can open an issue or a PR or otherwise waste a maintainer's time is broken. Fundamentally insecure under this type of attack. Now that the exploit is being used widely, and costing us immensely, we need to put a lid on it. If the only way to guarantee an AI bot (or its meatspace sock puppet) doesn't waste your time is to move to a "look but don't touch" model, then that's what we need to do. I think this would be a reasonable default:

Public repos are read only except for contributors who have been given specific permission, and those permissions are granular e.g. in order of increasing damage potential:

- comment on issue

- create issue

- comment on PR

- create PR

- run CI against PR

- etc.

In other words, shut it down.

> forget about the shutting it down and think of something actually realistic.

Why is it not realistic? Small teams do excellent work. Keep your team small and trusted. Only accept contributions from your team, and people outside your team who are personally vouched for by someone on your team. It's like climbing mountains or sailing or any other type of inherently risky activity--you don't go out with people you don't trust. It's eminently possible, you just don't like the idea of it.

Web3 is the closest analogue in recent memory, but if you go back further to the pre-enlightenment era (and some pockets of more recent history, particularly in isolated rural/colonial regions) you can see similar behaviors. It's mad religious fervor coupled with poor education. They see what their beliefs tell them they should see, and lack the mental rigor to analyze the actual data. Not their fault! It's our fault for letting them into the profession. Other disciplines are much better at keeping these folks outside the gates.

Interesting article, thanks for the link.

This capability-reliability gap (excellent term btw, more people need to think in these terms or we'll be in real trouble) is also infecting LLM assisted outputs. I just tried VSCode again tonight after a ~3yr hiatus and goddamn has it deteriorated. Lots of new features, lots of interesting looking plugins, but 3 out of the 5 plugins I tried for code CAD (the reason I downloaded VSCode again at all) were completely unusable--like couldn't even be made to work at all--and the other two didn't do anything like what they claimed. Also VSCode itself got into some kind of spastic loop trying to log me into github, and seemed incapable of recognizing the virtual environment in a python project's workspace... It also feels like the UI got even slower. This situation is bad.

> I expect soon we will see Rust forks with a pro-LLM policy

I sure hope so. I expect the end result will disprove the following:

> The Rust team will never be able to catch up to them

The AI jackasses have been braying in this key for going on a few years now, and there hasn't been one single time any of this breathless noise has resulted in something meaningfully superior. It's time to put up or shut up. Enough bullshit talk. If you can vibeslop a better Rust (or whatever), JFDI and leave everyone behind.

OK.. but what would you have all this "automation" actually do? What is Siri failing to do that you want it to do? How would customizing an assistant (for whatever definition) help?

> Something like this could help enable the "ever present" phone assistant, without scraping all your personal data and sending it off to Google/OpenAI/etc

OK so show me what that's for. Show me something useful you can do with that ability.

> Imagine if keywords in a chat would then trigger searches on your local data to bring up relevant notes/emails/documents into a cache, and then this cache directly powers your autocomplete (or just a sidebar that pops up with the most relevant information).

I'm really trying but.. idgi? I truly cannot imagine how this would improve my life in any way...

> Its cool. Enjoy it.

No. It sounds like a useless complication on my watch. I don't fucking care if it can tell me the phase of the moon. I can look up at the sky and see the moon and know what phase it is.

EDIT: You say:

> If you understand anything about the math and science behind LLMs, you'll understand that this is an achievement worthy of sharing to a community like HN.

OK. So educate me. Tell me what I'm missing.

> people make stuff to make stuff. There don't need to be product use cases.

OK. Great! So it doesn't need to be a commercial product. But does it do something (anything?) interesting? I'm interested in your games example, I'd love to see it done in real life. IIUC, game AIs are actually much more constrained and predictable for play-ability reasons. If you let it go all free form a plurality of players have a "WTF??!?" experience which is super Not Good.

TBH the 1999 visionary documentary The Wild Wild West is the foundational pedestal upon which I emplace technology writ large.

I get that, but I still can't imagine what it might be for. TBH I don't have a smart watch, because I can't think of anything I'd want one for--my mechanical watch keeps time to within a few seconds per month and the lume lasts all night. I don't know what making it "smarter" would do for me, it does an A+ job of being a watch. What are the things that "everyone" can build with this that actually matter? Like, what is the differentiator?

EDIT: To be clear, the monoculture of phone operating systems sucks. If this somehow enables more entrants into that space then I'm all for it. However, I don't see this in particular being the deciding factor... For example, the reason I don't run a 3rd party operating system on my phone isn't because it's lacking Siri or "OK Google" (if these things went away tomorrow I'd barely notice), it's because it would be a pain in the ass to make it be a phone.

I mean.. Siri basically works? When I'm driving I say "Hey Siri, find me a gas station along my route", and it does. Or I say "Hey Siri, call Joe Bob mobile" and it does. Or I say "Hey Siri, play me a podcast". This is kind of a solved problem already? When I'm driving this is literally as complicated of a distraction as I want--I'm not going to be dictating emails or texts. When I'm not driving, the touchscreen keyboard (as shitty an interface as that is) is 100x better than voiced natural language commands.