Yeah lemmy-ui doesn't escape/sanitizes properly...
philm
joined 2 years ago
That only really works, if the method is self-contained, and written in a language that GPT has seen often (such as python). I stopped using it, because for 1 in 10 successful tries I waste time for the other 9 tries...
This.
If I'm writing something slightly more complex, ChatGPT(4) is mostly failing.
If I'm writing complex code, I don't even get the idea of using ChatGPT, because I'm only getting disappointed, and in the end waste more time trying to "engineer" the prompt, only to get disappointed again.
I currently cannot imagine using ChatGPT for coding, I was excited in the beginning, and it's sometimes useful, but mostly not really for coding...
$0.7k annually? Is it anyhow possible to live with that low salary in India? I can't even live a month with that here, even if I don't buy anything but the cheapest food and live in the smallest apartments here...
I think the crux of good software design, is having a good feeling/intuition when to abstract and compose, and when to "inline" code as it's used only once anyway (in which case "linear code" is IMHO often the better choice, since you're staying close to the logic all the time without having to jump between files etc.).
I agree the examples of the google blog are bad, they don't really show how to compose code nicely and in a logical way (whether that would be more data-oriented, imperative or OOP).
Yes
It could definitely use integration with a linter so it doesn’t generate subtle bugs around generative naming mismatching actual methods/variables, but it’s become remarkably good, particularly in the past few weeks.
Maybe I should try it again, I doubt thought that it really helps me, I'm a fast typer, and I don't like to be interrupted by something wrong all the time (or not really useful) when I have a creative phase (a good LSP like rust-analyzer seems to be a sweet spot I think). And something like copilot seems to just confuse me all the time, either by showing plain wrong stuff, or something like: what does it want? ahh makes sense -> why this way, that way is better (then writing instead how I would've done it), so I'll just skip that part for more complex stuff at least.
But it would be interesting how it may look like with code that's a little bit less exotic/living on the edge of the language. Like typical frontend or backend stuff.
In what context are you using it, that it provides good results?
I would actually encourage it to be extremely verbose with comments
Yeah I don't know, I'm not writing the code to feed it to an LLM, I like to write it for humans, with good function doc (for humans), I hope that an LLM is smart enough at some day to get the context. And that may be soon enough, but til then, I don't see a real benefit of LLMs for code (other than (imprecise) boilerplate generators).
Yeah my thinking as well.
Addtionally, why I think other system language competitors like Zig or Nim aren't succeeding long-term, is because of fast growth and already big ecosystem of Rust. Zig may be better though for some use-cases (when you want to avoid all the mental overhead, and the application stays simple).
On another look, though, we have to keep in mind, though that this is code-golf, so in no way representative for actual code-bases.
Hmm interesting, I would've thought that Haskell would rank much higher
Has it though?
AFAIK it has like 4k tokens which roughly translate to 3k words or something like that.
The API has 8k or even 32k (this model should be interesting for something like that, unfortunately still don't have access to it...)
So the docs have to be rather short, that it gets all the relevant stuff, or am I missing something?
At this point, I think it's almost mainstream, and it's still growing fast (and it's getting better, rust-analyzer is really awesome these days, I was there at the beginning, no comparison to today...))
I may be biased, but I think it'll be the next big main language probably leaving other very popular ones behind it in the coming decade (Entry barrier and ease of use got much better over the last couple years, and the future sounds exciting with stuff like this)