stephc_int13

I fully agree with that.

Past experience has shown what while Turing complete config is powerful, it is also a double edged sword and I would be wary of using it.

I admit I was a little bit curious about DOGE and its outcome, as a software guy I've seen huge amount of bureaucratic inefficiencies, sometimes wondering how some companies could survive wasting all that money for nothing.

I've also often be irritated by the slow and sometimes absurd processes of French administration, and I am pretty sure the whole thing is far from running as efficiently as it could.

That said, I gave way too much credence to Musk and his clique, turned out the efficiency theatre was nothing more than a smoke screen to cover a different style of operation.

I am not sure they even tried to cut waste in a meaningful way, or if they had the competence to do so.

Zuckerberg was barely adult when he started Facebook. And he probably bumped into a few older guys who thought they knew better than him, and history proved them wrong.

He likely developed some irrational belief that clever and young beats anything else, and saw an echo of his own bravado in Alexandr Wang.

Turns out his heuristics were not calibrated properly.

It is the same reasoning as with the regular Decimation. It is all about disciplining their employees.

And it works.

We're not saying many ex or current Meta employees talking about their experiences here, even if I am sure that HN is pretty popular among this crowd.

And of course this is not unique to Zuck/Meta. We don't hear much from people working for Musk either.

He is one of the well known netcode guy.

But from the few interactions I had with him I would say he is quite abrasive, stubborn and probably somewhat on the spectrum.

But there is a special kind of unpleasantness in writing/debugging netcode for large projects, I don't think you can be agreeable and still you your job correctly.

The trajectory of id software is interesting.

The company was successful, had one of the most prestigious brand in the game industry, was early enough to capitalize on the rise of PC gaming, incredible talents and tech.

Yet it didn't transform into a Blizzard or Epic.

And it seems that both the early success and stall were the responsibility of one very talented but somewhat obtuse nerd.

I think the rationale is that they are already using typical car factory automation, but they see a huge potential market for general purpose robotics in the coming decades, they don't need the humanoids, they are simply dogfooding a future product.

I think this is smart and not very risky. Tesla is playing a similar game with Optimus, for now Hyundai/Boston Dynamics is at least 5 years ahead.

I am not one of them but quite a few programmers prefer not having to use the mouse at all when working.

The terminal is an old but astonishingly powerful user interface that is still evolving.

Good terminals can be very snappy and configurable in ways that most GUI are not.

There is also arguably an aesthetic/fetishism appeal to it.

I've worked in the terminal at some point of my career, as there was not many other choices, and I understand how someone can get really used to it.

It can be used for many things. But the main use is reverse engineering.

I tried it a few times and was not convinced by the autocomplete.

I found it less effective than free copilot autocomplete on vanilla VSCode.

If you think about it states are not much different than corporations.

Taxes is human labor dedicated to the maintenance of state.

I think it is naive to think that once humans are not needed they will magically be fed by the machines.

If you own a factory, what you see is a big box with input and output, human labor is just an annoying variable, removing it won't turn the entire thing "free".

The owner will still expect to trade the output for something else for his benefit.

You missed the point entirely.

What fuels the economy, for now, is human work. Being a consumer simply means that you are heavily incentivised to take part (by working).

If machines can do the same or better, whoever owns them will end-up seeing unproductive humans as dangerous and wasteful bloat.

It won’t be a life of infinite leisure if you don’t own machines.

Otherwise you could say that humans are working for free as well, as they can cultivate their own food and even build themselves by making babies.

Of course, but none of them is working for free, each consume energy and materials.

It is never really “for free”.

It appears free because machines, like perfect slaves, don’t ask to be paid for work.

They still need to be fed though, energy, iron, etc.

No matter how you turn it, it can’t be free. Meaning you can’t benefit from it without owning it or trading something.

I genuinely have no idea how that transition would happen.

And I agree that it would pose many unforeseen challenges.

This is why the transition is the interesting part, not the sci-fi end game with a world populated with billions of robots doing everything.

No, you don’t feed machines with bitcoins or any kind of currency, you feed them with Joules, Iron, Copper, rare earths etc.

Yes, the transition is unlikely to be linear and without conflict, if this was ever possible. But I am sure that some would be happy to control armies of bots and very few humans.

I have not, but I am curious.

I think the end-state is not that interesting, but the transition could not happen overnight and seems both difficult technically and would be unlikely to happen without a fight.

The idea of a consumer based economy has always appeared dumb to me.

The reason why the masses should consume is to motivate them to work. And the reason why having a large amount of people working is that human work has been producing a surplus basically since the dawn of civilization.

This surplus is partially shared but tend also to "trickle up", contrary to some weird beliefs, as can clearly be seen almost everywhere you look.

But if you imagine a sci-fi world where machines can build and do everything humans can do, the concept of a human-centric economy would be pointless.

Machines don't need to be motivated to work, they just need energy, materials and obeying to whoever controls them.

This kind of economy would be less abstract and more directly related to physics.

The issue with billionaires is that some of them got insanely rich while being net negative from a societal perspective.

The most famous ones ended-up in prison (Sam Bankman Fried, Elizabeth Holmes, Jeffrey Epstein, Bernie Madoff) but anyone with a basic grasp of statistics and criminal behavior know that many others will escape the justice system forever.

It does not mean that all billionaires are bad, the criminals are not the majority, but there are enough criminals to justify skepticism and scrutiny.

The merge could be named Musk Industries, quite fitting considering his hubris and association with comic books.

Not the same solar panels. As with most equipment, the constraints are very different in space/orbit, and the suppliers are not the same.

They probably have some kind of fallback system, but the visual-inertial odometry they are using for spatial positioning that is working pretty well when stationary (at home) tend to break badly on a train or plane.

His rhetoric is a bit obsessive and frankly biased against AI.

That said, I think his voice is useful as a counter to the mainstream opinion.

Given the amount of investments, approaching AI from the angle of economics seems correct.

We all have some level of personal experience using AI/LLMs, both chatbots and coding tools, and I personally enjoy using them, but I am sure this experience is relevant in this discussion.

I also enjoy luxury hotels, gourmet food, jet skis and helicopters, but this is not something I indulge in often because of the cost-utility ratio.

The real cost of AI may or may not be lower than its utility. The bet is that utility is increasing while cost is falling.

I agree about the C stdlib being outdated and the worst part of C. This is why I don't use it.

A new modern C standard library would be a very good thing. But I am ok writing my own, tailored for my own needs and style.

Linters are not exactly a new idea, and they can be pushed quite far.

I am generally against the idea of language support for meta-programming, even using C macros or C++ templates.

I do a lot of meta-programming/codegen with external tooling.

I know, and I don't see unused return values as an error, it should be a warning at best.

And the compiler should be just a bit smarter to avoid the .{} thing when it is not strictly necessary.

Lindy effect is very strong in the programming languages world.

Zig is a very interesting and well built language, but it is also niche, unstable and not offering radical improvements over C.

C, as the foundation of all modern kernels and system tools, among many other things, is extremely unlikely to go away soon.

Everything you do in Zig can be done almost in the same way in C or C++.

I would say the real value of Zig is in the standard library that is more modern and arguably better.