News for hackers. Stuff that matters.
Re:AI is impressive because we've failed at personal computing by BobaFloutist on Friday August 08, 2025 @12:33PM (#44838941)
Wait, which is the correct programming language to defend? C and Javascript are on pretty opposite sides of most spectra....
Re:AI is impressive because we've failed at personal computing by mamcx on Friday August 08, 2025 @02:53PM (#44840388)
Today is easy (finally we have Rust, Zig, Odin, Swift, Go etc that show marked improvements), but the op is correct: A lot of progress was stalled because people defend sub optimal tools.

Every time somebody suggest "but C is wrong" the answer was "You should not do the wrong, be careful every time and things will be fine!"

P.D: Yesterday we have pascal/ADA/effiel/ocalm and others, but the major issues is that C/JS should have be improved to remove the bad things, and add the good things. Is not as if what or why to do it was a mystery, it was just the fear and arrogance against change.

And this cause too much inertia against "reinvent the wheel" and too much hate against anybody that even try

Re:AI is impressive because we've failed at personal computing by jeltz on Friday August 08, 2025 @01:00PM (#44839263)
Languages which are not hard to use and dangerous. I work in C professionally right now but would much rather work in either Zig or Rust.
Re:AI is impressive because we've failed at personal computing by oivey on Friday August 08, 2025 @02:03PM (#44839835)
He seems to have mistaken his personal opinions on which languages and language features are good for some objective truth.

Ironically, that’s part of why we can’t have nice things. People who aren’t open to other viewpoints and refuse to compromise when possible impede progress.

Re:AI is impressive because we've failed at personal computing by ajkjk on Friday August 08, 2025 @07:50PM (#44842878)
well I was being flippant

but

what do we have other than our opinions? I think everything sucks so that's what I said.

Re:AI is impressive because we've failed at personal computing by ajkjk on Friday August 08, 2025 @12:49PM (#44839143)
the new ones we should have made by now that fix their problems...

Zig and Rust and TS are a start

Re:AI is impressive because we've failed at personal computing by xigoi on Friday August 08, 2025 @01:29PM (#44839548)
They are on the same side of the “number of serious design flaws due to historical incidents” spectrum.
Re:AI is impressive because we've failed at personal computing by BobaFloutist on Friday August 08, 2025 @02:05PM (#44839855)
Right, so, to reiterate the question, which language(s) are on the opposite side of that spectrum?
Re:AI is impressive because we've failed at personal computing by dmurray on Friday August 08, 2025 @12:41PM (#44839035)
Presumably the spectrum that would have Rust and Typescript at the other end.
Re:AI is impressive because we've failed at personal computing by BobaFloutist on Friday August 08, 2025 @02:05PM (#44839857)
Is that just age?
Re:AI is impressive because we've failed at personal computing by klardotsh on Friday August 08, 2025 @02:15PM (#44839961)
Well, sure, age is part of it. I would hope languages coming out 40-50 years after their predecessors (in the case of Rust following C/C++) would have improved upon those predecessors and learned from the ideas of computer science in the intermediate years.

(Coincidentally this is one of my chief complaints about Go: despite being about the same age as Rust, it took the avenue of discarding quite a lot of advancements in programming language theory and ergonomics since C)

Re:AI is impressive because we've failed at personal computing by cestith on Friday August 08, 2025 @02:59PM (#44840454)
Go has a much different set of design goals than Zig, Nim, or especially Rust. Go is really for people who want a few major improvements on C like easier builds, faster builds, higher-level standard string handling, direct support for some form of concurrency, an optional GC which defaults to on, and a few syntax tweaks - things that a modern C might have that were not going to make it into a standards revision. Rust, to support ownership and the borrow checker at compile time, had to build a useful language around that hugely helpful but quite restrictive requirement. They subsequently went different directions than the Go team on a lot of the other language features. Zig, Nim, and D are something in between those extremes in their own ways.

As someone with a background of a lot of time with Perl and the Pascal/Ada family who was rooting for Julia, Go strikes a good balance for me where I would have used Perl, Python, Julia, Ruby, or shell for system tasks. I haven’t done a lot of work in Rust yet, because in the infrastructure space speed to implement is usually more important than performance and Go is a lot faster already than Python or Ruby and because my team as a whole has much more experience with it. I certainly appreciate why the folks at work who do the performance-critical network and video software use mostly Rust to do what they used to do in C, Java, or C++.

Re:AI is impressive because we've failed at personal computing by tempfile on Friday August 08, 2025 @01:03PM (#44839294)
Programming languages are not on a spectrum. I guess they dislike the syntax of JS and the low-level no-batteries of C.
Re:AI is impressive because we've failed at personal computing by BobaFloutist on Friday August 08, 2025 @01:21PM (#44839467)
Programming languages, like literally everything else, are on many spectra.
Re:AI is impressive because we've failed at personal computing by tempfile on Friday August 08, 2025 @01:40PM (#44839645)
Did your comment always say "most spectra"? I swear it was "a spectra" before. Sorry, I must have misread.
Re:AI is impressive because we've failed at personal computing by BobaFloutist on Friday August 08, 2025 @02:04PM (#44839844)
It did, but it is an unusual plural to encounter in conversation so I don't blame you for missing it. Kudos for owning up!