graydon points in that direction, but since you're here: how feasible is a hypothetical Fil-Unsafe-Rust? would you need to compile the whole program in Fil-Rust to get the benefits of Fil-Unsafe-Rust?
It's reasonably easy if you can treat the Safe Rust and Fil-Unsafe-Rust code as accessing different address spaces (in the C programming sense of "a broad subset of memory that a pointer is limited to", not the general OS/hardware sense), since that's essentially what the bespoke Fil-C ABI amounts to in the first place. Which of course is not really a good fit for every use of Unsafe Rust, but might suffice for some of them.
I think the point is that Fil-C makes programs crash which didn't crash before because use-after-free didn't trigger a segfault. If anything, I'd cite Redis as an example that you can build a safe C program if you go above and beyond in engineering effort... most software doesn't, sadly.
Redis uses a whole lot of fiddly data structures that turn out to involve massive amounts of unsafe code even in Rust. You'd need to use something like Frama-C to really prove it safe beyond reasonable doubt. (Or the Rust equivalents that are currently in the works, and being used in an Amazon-funded effort to meticulously prove soundness of the unsafe code in libstd.) Compiling it using Fil-C is a nice academic exercise but not really helpful, since the whole point of those custom data structures is peak performance.
I imagine they can spot it pretty quick using machine learning to spot unlikely API access patterns. They're an AI research company after all, spotting patterns is very much in their wheelhouse.
If you can get a specialized agent to work in its domain at 10% parameters of a foundation model, you can feasibly run locally, which opens up e.g. offline use cases.
Personally I’d absolutely buy an LLM in a box which I could connect to my home assistant via usb.
Can you (or someone else) explain how to do that? How much does it typically cost to create a specialized agents that uses a local model? I thought it was expensive?
An agent is just a program which invokes a model in a loop, adding resources like files to the context etc. It's easy to write such a program and it costs nothing, all the compute cost is in the LLM call. What parent was referring to most likely is fine-tuning a smaller model which can run locally, specialized for whatever task. Since it's fine-tuned for that particular task, the hope is that it will be able to perform as well as a general purpose frontier model at a fraction of the compute cost (and locally, hence privately as well).
Back in my day, we lamented the loss of bang paths for email... and you had to pay Robert Elz to bring in a news group because munnari connected Australia to the world...
You’ll likely need to wait a couple decades for this. Kids need to not spend all their free time helping on the farm first and their currency has to appreciate such that they can afford a computing device that isn’t a toy. With yields seeing such big increases it looks feasible, but we’re still early.
TUIs being designed by engineers for engineers make them rather timeless. Extra points for being keyboard-first: lots of modern GUI tools don’t even consider the keyboard for anything other than text input, to the point that even tab order is broken, if it works at all, or the escape key closes multiple stacked modal windows, or enter doesn’t submit the dialog, or…
Agree - I am an ardent capitalist, but a conscious capitalist. I believe the purpose of capitalism redirected can be used as a vehicle for massively changing economies and lives - such as in this case.
It's not capitalism, it's technology. That can often go together with capitalism, but Russia from 1917-60 and China from 1960-2025, say, are big counter-examples. As are the many poor countries with capitalist economies. Growth in electrification, transport infrastructure, manufacturing and mechanized agriculture will grow any economy, capitalist or socialist
Strongly disagree, you're example is nonsensical as it's normally used to prove the exact opposite. Nearly every quality of life improvement and economic boom in China and Russia during those periods are directly tied to adopting some parts of capitalistic systems.
Those that happened in the USSR and China, no? After the start of electrification and active mechanization of agriculture, more peasants died of hunger there than in the previous 100 years (in Russia more than in the previous 200 years).
That is simply how socialism works. Land as a mean of production is no different from a factory, so naturally, the products created on the land belong not to the peasants (Petite Bourgeoisie), but to society as a whole.
And guess what this bourgeoisie did when they found out that the grain they produced would now become common property (they sharply reduced the amount they producing).
> but Russia from 1917-60 and China from 1960-2025, say, are big counter-examples.
Russia and China are good examples of that.
We have Khrushchev's memoirs about how, before the communist revolution, he, as a simple worker, lived better than workers live 40 years after the revolution. That is, the period from 1917 to 1960 in the USSR was one of complete stagnation, despite all the technological progress.
And in the example of China in the second half of the 20th century, we see yet another confirmation: their standard of living was literally directly proportional to the level of implementation of capitalist mechanisms.
> As are the many poor countries with capitalist economies.
As far as I understand, there is not a single poor capitalist country. Name a single poor country where private property is reliably protected and people enjoy economic freedom. There is no such country. As soon as even the poorest country begins to protect private property and guarantee economic freedom, it becomes rich within 10 years or something.
You make some reasonable points. But you're wrong to discount technology. Consider what Russia and China would have looked like if they hadn't electrified and mechanized.
"PC Mode" or "Gaming mode" or whatever is necessary - I can tell any other mode easily just by moving the mouse, the few frames of lag kill me inside. Fortunately all tvs made in this decade should have one.
graydon points in that direction, but since you're here: how feasible is a hypothetical Fil-Unsafe-Rust? would you need to compile the whole program in Fil-Rust to get the benefits of Fil-Unsafe-Rust?
reply