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

We Are the Last People Who Know How It Works

unix.foo

Post reply on HN
Print view View original

251–260 of 283 posts

Re: We Are the Last People Who Know How It Works

#251

Post by ranger_danger »

budman1 wrote: post #156
vitally3643 wrote: post #104

Earlier quoted context omitted.

There was a period of a century or more where steam engine design parameters were determined by experiment and gut feel. The physics of fluid dynamics required to explain why a design is optimal just did not exist yet. There's a lot of places in history where engineering far outpaced the science required to explain it.

If your design work is not based upon science and mathematics, it is not engineering. Not saying engineering is required to build or improve a product; it is not. "try this" is not engineering.

> "try this" is not engineering.

It was always engineering. You're making the assumption that their theories are not rooted in anything coherent at all, but almost always this is not the case. Usually a "what if we..." thought is based on other knowledge, and there's absolutely nothing wrong with trial and error... this is what research and experimentation, areas of science, are all about.

Re: We Are the Last People Who Know How It Works

#252

Post by c7b »

matheusmoreira wrote: post #213
c7b wrote: post #83

Moral panic. We all have private tutors on every subject at our disposal now. It takes some special kind of mental gymnastics to conclude from this that no one will learn anything anymore henceforth. The article is just a long-winded way of saying "Kids today have it too easy", or, equivalently, "I'm getting old".

Yeah, I'm learning more than I have ever learned before. I suppose the worry is that people won't want to learn, though. That people will prompt the AI to just do things without caring about why and how they are done. That's a legitimate worry, but it's not new.

Desire to learn is deeply ingrained in humans and hard to root out. The education system does a pretty good job at it, unfortunately, but nothing ever is 100% effective.

Re: We Are the Last People Who Know How It Works

#253

Post by stevage »

>The modem sang its negotiation out loud, and that shriek was two machines arguing about how to speak to each other, and after enough times you could hear when the argument was going south and the call was about to drop. You set little jumpers on drives with your fingernail. You knew which interrupt your sound card answered to, because if you guessed wrong it answered to nothing. The machine was made of edges and you cut yourself on them, and that is how you found out where they were.

Wow. I had literally all of these experiences. I remember how you used to have to remember the Sound Blaster interrupt settings and pass them to any game that used it. And I'd completely forgotten about IDE jumpers, but yep - how fiddly it was when you wanted to put 4 hard drives in the same computer, figure out which were slaves and masters, etc etc.

Re: We Are the Last People Who Know How It Works

#254

Post by zombot »

footydude wrote: post #53
apsurd wrote: post #32

Earlier quoted context omitted.

The financialization of everything is what's ruining everything. In the computer and internet realm, there's warm nostalgia from hours spent tinkering, building one's own PC, reformatting C drives due to malware, searching for "snippets" to add a forum or pimp a myspace page. But inevitably the money incentives come to dominate. It's all of our doing, we all dream of a better life to put it charitably. Now everything…

In my mind, 'tinkering for fun and knowledge' was never meant to be profitable - if you're tinkering with the express aim of making a profit i'm not sure you're really 'tinkering' so much as you're trying to create a product/service/output for someone else to pay for? There's absolutely nothing wrong with doing that - I'm just not sure the 'ethos' of tinkering has anything to do with trying to make money and is usual…

> 'tinkering for fun and knowledge' was never meant to be profitable

That's why it's "for fun and knowledge" and not "for profit".

Re: We Are the Last People Who Know How It Works

#255

Post by beej71 »

The thing that freaks me out about "the end of programmers" is that we're going to introduce a knowledge gap. And I'm unaware of any knowledge gaps in anything else we do. For any technology we use, we have humans who can work together to build it from dirt.

When no one knows how to program the tools that we rely on to fly our passenger planes, drive our cars, and manage our bank accounts, what happens when the LLM gets it wrong?

Re: We Are the Last People Who Know How It Works

#257

Post by archagon »

staindk wrote: post #81
CommieBobDole wrote: post #22

The issue with this is that we don't know how it works. Generally speaking, we know how the level of abstraction that we were born with works. We might have some understanding of one or two previous levels, but that decreases the farther down you go. We might understand the next level, and some of the next after that, but eventually people will be making things that we don't have the context to understand without hav…

I do think there's something to be said for the big difference in the current abstraction layer jump - the loss of determinism. It is meaningfully different.

More than that: it's not an abstraction layer jump, because AI is not an abstraction.

Re: We Are the Last People Who Know How It Works

#258

Post by JumpCrisscross »

mrob wrote: post #107
JumpCrisscross wrote: post #88

Earlier quoted context omitted.

Plenty of physical (and biological) processes are subjectively non-deterministic outside clean rooms. Doesn’t mean our ancestors couldn’t forge and selectively breed.

There's a big difference between "deterministic + noise" non-determinism and "intelligent agent" non-determinism. Only the former can be statistically modeled to characterize and work around the noise with any reliability. If you measure hallucination rate for one prompt it tells you nothing about hallucination rate for another prompt.

> There's a big difference between "deterministic + noise" non-determinism and "intelligent agent" non-determinism

I think I agree, but could you expand?

My point is humans have made progress on understanding systems without clean determinism. Genetics and cosmology is technically deterministic, practically speaking, but I don’t think that’s accessible to pre-modern societies. Instead, trends and tendencies were observed, acted and capitalized on.

Re: We Are the Last People Who Know How It Works

#259

Post by globular-toast »

jfengel wrote: post #139
globular-toast wrote: post #123

Earlier quoted context omitted.

A person dies every 26 seconds in a road traffic incident somewhere in the world. A big part of that is people using machines they do not understand.

In what sense do they not understand it? What could we teach them that they don't already know?

They don't understand the dangers involved because the soft suspension and airtight cabin shield them from it. They don't have an intuitive sense of speed, force, momentum, traction etc. because they never rode a bicycle and fell off.

Re: We Are the Last People Who Know How It Works

#260

Post by JumpCrisscross »

eikenberry wrote: post #221
tines wrote: post #155

Earlier quoted context omitted.

People use the word "determinism" when they really mean something akin to "linearity", i.e. the predictability of a change in input on a change in output. Compilers for example are both deterministic and "more linear" in the sense that I can tell what the output will look like given a change in input (yes yes optimizations violate this to a small degree). LLMs can be made totally deterministic, but a seemingly insign…

The generally used term that I'm familiar with for this is "referential transparency", that given the same inputs you'll get the same outputs every time. LLMs can be deterministic (referentially transparent) but almost none are, i.e. when they are given the exact same input they do not return the exact same output.

Wait, what is the light between that definition of referential transparency and determinism?
Post reply on HN