gf000

How to properly secure it?

If you put it in a sandbox than herdr can no longer see it, can it?

Are they working together with interdependencies or not?

E.g. explorer.exe is not replaceable on windows so the display server and that is one software. (The kernel is NOT though, because it can surely run in a headless way). It's not that hard of a concept..

Well, does the way spreadsheet cells are handled has any relevance to presentations? If not, then they don't increase each other's complexity budget.

systemd-init has nothing to do with network resolution.

How does go even enter this discussion? It's closer to JS than to zig/rust.

> if you understand how the CPU, memory, network, and disk are processing

I don't think it's possible in the general case, it's way too complex for that. Sure for trivial examples it's trivial to claim one program is faster than the other, but can you really claim that your special allocator used in one place plus the extra argument used in calling that function is worth it without writing both options and measuring it? Like even measuring it is non-trivial, will it be inlined at the use site, how is it called, etc.

Well, my only gripe with it is that it's C. Otherwise it's very modular.

Sysvinit is not 1000 lines if you add all the scripts you have to write to get feature parity with systemd. And now you have written that in the worst programming language ever created, bash, that is basically impossible to not be chock full of errors.

Frankly the amount of crying I have seen about systemd... and it's basically the most stable part of my Linux installs. Even though I really have seen everything else "catch on fire".

Why not go with something like graalpy first? That's much less risk in terms of "we end up slightly changing behavior"

Even better are the bank websites that refuse to work on multiple tabs, logging you completely out after detecting that..

Just simply ask for that, and they usually know what to do.

They should write a test and stash the fix they just wrote. In this state it should fail. Once they re-apply the patch, it should become green and this gave use enough information to deduce that the test does test something.

Well, banking is digitalized for a long time (such a long time that they actually are using mainframes and cobol), and it's definitely not the epitome of good software, so I think it was always important, maybe we were just less directly reliant on it.

From another comment:

> 10^87 variables, a point where the proposed new method surpasses the already existing sat solvers

Yeah, it's not efficient in the ordinary sense.

Ask for red-green tests. Without that they are often indeed useless.

"hey Claude, rewrite it in rust"

I'm semi-sarcastic here, but this is actually possible now, while it was an often futile and very expensive roundtrip before AI resulting in two versions of your software that you have to simultaneously maintain.

Yeah shit like this is/was always extremely common on any kind of app created in the last 30 years. It's not related to vibe coding.

Bit didn't it also feel bad back then, you just forgot about it and has a rose-tinted view of the past?

Like software used to crash, were single-threaded and constantly frozen. How many times did you have to issue Ctrl+alt+del vs today? Even UIs were constantly out of sync with the underlying data, I would even go as far to claim that reactive frameworks like react cut down a big category of software bugs.

You seem to give too much credit for these companies and too less credit for governments.

First of all, most companies just want to make money, with minimal risk. It's significantly easier to make money when you are at worst doing grey area stuff and stay mostly legal (or even "better" is to lobby your thing into being legal). And any info known by more than 2 people is no longer a secret, and you ain't paying enough to your employees to lie under oath for their employer. So no, I don't believe it's happening all too often, and even if it does happen it's mostly out of ignorance, not a deliberate attempt to go against the law.

Though at this time it produces pretty bad text which is hardly useful for anything.

You can check what data is getting collected (e.g. decompile their distributed app at different versions), and you can - as a government - actually just go to their servers and check what they are doing (with necessary permits and lawful action), or even mandate them to show the source code.

Standing for an evil heinous party doesn't get offset by this, though, were it the case.

DNA is so many layers below human cognition, it's a bit like explaining the Moon landing's technology by a stick burning.

There is absolutely no DNA recombination during a thought process.

> leading them to suggest that LLMs will hit a wall in mathematical capability just beyond the convex hull of human mathematics

I mean, even just that would prove to be a very productive workhorse - potentially being able to automate away almost every white collar job.

> formal proofs are, as Rust provides something very similar to that

Since when? Rust just has an ML-like type system, and seemingly many have met that for the first time with Rust, even though it has decadeS of history, and multiple production languages running it just fine (ocaml, scala, etc).

Formal proofs are way way above an ML type system. Like how do you encode mathematical statements in Rust's type system? What about statements about program semantics, like program A is equivalent to program B <- in formal verification THAt is the type of an expression, and the implementation of it is the proof.

CS theory-wise, yes. Practically, no.

Memcpy of what? I'm fairly sure Firefox's internal representation is different from Chrome's, so you just replaced parsing from X with parsing from Y.

That's probably the very reason it's arguably not the best language ever created. It has way too large surface area, getting close to C++ levels.

Sure, it's a valid language design, but it's also a negative in "understandability of each line of code", and thus maintainability, and even stuff like tooling, verification.

I don't think it's a fair comparison. Maybe against deepseek pro would be more similar in terms of model size

I don't think it would be all that useful, give the very way it works (modulating motor speed) - your rotation would not be consistent in speed during even a single turn, which makes exposure times different each frame, and possibly too low.

Not disagreeing you, but can we please stop using tiobe? It's by far the worst metric that doesn't mean anything (even though it is correct here, but so is a broken clock twice a day)

Scala has refined implicits into givens and extension methods.