pron

77 ms·

871.
by pron 1y ago
> This is the future. It is inevitable. "This" does a lot of unjustifiable work here. "This" refers to your successful experience which, I assume, involved a program no larger than a few tens of thousands lines of cod
872.
by pron 1y ago
> Offloading some of that to AI lets one free some mental capacity for work that actually benefits from that. Maybe, but I don't feel (of course, I could be wrong) that doing boring tasks take away any mental capacity; they feel m
873.
by pron 1y ago
In my case, there is no wrong or impossible direction, just a technical detail that you realise you must overcome when you start to code and that I doubt the model will be able to solve on its own. What it should do is start coding, realise
874.
by pron 1y ago
But the questions I'm interested in cannot be asked until the programmer starts to code. It's not that the task is unclear, but that coding reveals important subtleties.
875.
by pron 1y ago
Here's the thing, though. When working with a human programmer, I'm not interested in their code and I certainly don't want to see it, let alone carefully review it (at least not in the early stages, when the design is likely
876.
by pron 1y ago
I've yet to use an LLM for coding, so let me ask you a question. The other day I had to write some presumably boring serialization code, and I thought, hmm, I could probably describe the approach I want to take faster than writing the
877.
by pron 1y ago
The compiler was much faster, the tooling better, and it was easier to find knowledgeable programmers (we were spending quite a bit of time sifting through thick Ada reference manuals). Whatever correctness benefits Ada provided at the lang
878.
by pron 1y ago
Algol 68 and Pascal weren't memory-safe, and as for Lisp, Smalltalk, and ML, their style of memory safety - based on GC - took over the world pretty much the second it became practical enough for widespread use. It is true that some de
879.
by pron 1y ago
The idea that memory safety is a binary choice between what Rust provides and anything less than that has absolutely no grounding in either empirical or theoretical results. For example, it is true that there are good empirical reports that
880.
by pron 1y ago
The two languages have completely different design philosophies and a completely different feel leading to a completely different experience. I would say they're more different from each other than Java is different from Python. Some l
881.
by pron 1y ago
Because fostering the growth of tech companies isn't the most important thing, and the European bureaucracy has been more effective than the American one at more important things, including reducing the harm of Big Tech. I'm not o
882.
by pron 1y ago
> And if it failed to do that, I am confident that the problem is a mistake in the logic theory proof. You can axiomatically add oracles without introducing logical inconsistencies. It is commonly done in complexity theory. The "mis
883.
by pron 1y ago
> I have absolutely no idea how physical reality can contradict a statement about a formal symbol game. Of course it can! For example, if some logic theory contains the theorem that a particular realisable polynomial-time algorithm that
884.
by pron 1y ago
> Instead, at worst, he would simply move it from the category of being a real formula, to being an ideal formula. No. No result, either ideal or real, may contradict reality (it's just that since infinitary statements do not desc
885.
by pron 1y ago
> Some of the humans who have thought about it do reject them. I think they reject them only if they misrepresent Hilbert's formalism, because formalism does not assign them any meaning of truth beyond the symbolic. It makes no st
886.
by pron 1y ago
I can't locate my Heijenoort right now, but here's a description from the Stanford Encyclopedia of Philosophy [1] (which points to Heijenoort): The analogy with physics is striking... In the second half of the 1920s, Hilbert repl
887.
by pron 1y ago
> In what sense can a finite set exist and be finite when it is unfindable, unverifiable, and has unboundable size? In the same sense that we could say that every computer program must either eventually terminate or never terminate witho
888.
by pron 1y ago
There's no appeal to emotion here. They're saying that people who had an outsized influence in electing this administration have an outsized responsibility for its outcomes. Also, people dying or getting ill isn't an emotion.
889.
by pron 1y ago
> But it turns out that a C compiler will give you the "perfect" code directly while the comptime Zig version is fairly complicated. In this case both would (or could) give the "perfect" code without any explicit c
890.
by pron 1y ago
C also creates a distinction between compile-time and run-time, which is more arcane and complicated than that of Zig's, and your code uses it, too: macros (and other pre-processor programming). And there are other distinctions that ar
891.
by pron 1y ago
You can think about a TLA+ spec as the level of design between the "natural language design" you have in your head or on paper and the code. It makes it easier to go from the idea to the code, and allows you to explore ideas, at a
892.
by pron 1y ago
First, this is true in any architecture. Second, in Java you can do just that with the use of semaphores guarding different resources with a different number of leases. The one resource you can't easily schedule well is CPU, but that&#
893.
by pron 1y ago
The CPU is only contended once it's at 100%. I am not aware of any interactive server (as opposed to batch processing) that behaves well at 100% CPU (beyond short bursts), so I think that the hope you can actually balance the CPU using
894.
by pron 1y ago
What you usually want to limit isn't the number of threads but the number of threads doing a particular operation, such as accessing some service, so it's more flexible (and efficient) to just guard that particular operation with
895.
by pron 1y ago
It should be the other way around. The concurrency-limited resource is usually accessed by IO (some service), so a thread would wait and then acquire the semaphore before doing the IO.
896.
by pron 1y ago
> By switching to virtual threads, you traded off the safeguards once provided by your (bounded!) thread pool and OS scheduler. The OS scheduler provides no safeguards here (you can use the thread-per-task executor with OS threads - you&
897.
Two AIs Compete at Variant Sudoku [video](youtube.com)
1 points by pron 1y ago|0 comments
898.
by pron 1y ago
When I've seen examples of such cases, my impression was that they could be improved by taking the AI out of the picture and using some "low code" solution.
899.
by pron 1y ago
The funny thing is that, on a purely technical level, LLMs are more likely to do a better job at replacing upper management than replacing developers. If companies really want to save money, they should let AI replace the CEO.
900.
by pron 1y ago
Sure, but do you think that if nothing had ever happened in Gaza, the Trump administration wouldn't have found some other pretext to go after higher education, and foreign students in particular? They're defunding research program

More ›