You will always be better at decisions than an n-dimensional matrix of numbers on an overpriced GPU.

I’d be careful about these claims. Maybe with our current iteration of “attention-based” LLMs, yes. But keep in mind that our way of processing information is strongly limited compared to how much data is fed to these LLMs while training, so they in theory have a lot more foundation to be able to reason about new problems.

We’re vastly more capable at the moment at interpreting our limited view on foreign code, being actually creative, find new ways to reason, yes. Capable developers (open source…) often have seen quite a bit more code than the average developer and are highly skilled, still with just a tiny subset of the code that an LLM has seen.

But say these models improve in creativity and “higher-level of thought” through whatever means (e.g. through more reinforcement learning). Well, let’s just say I’m careful with these claims. These LLMs are already quite a help with stupid boilerplaty code (less so with novel stuff, and writing idiomatic non-redundant code, but compared to 2-3 years ago it’s quite a step already, to the point that they’re actually helpful, disregarding all the hype and obvious marketing strategies of these AI-companies)

Helix:

because the massive ecosystem of JS components makes you more productive.

Slightly less ironic: I question even this right now (as I have to suffer from endless “hot”-reloading and browser-crashes because of Next.js bloat).

I think the massive ecosystem has fewer high quality libraries than Rust at this point. I use both JS/TS in frontend and Rust (either frontend more as a hobby and backend) extensively, and I very often check the dependencies-source, and even more often rewrite it (unfortunately not in Rust), because of low-quality. And it’s sooo slow… the tooling and the frontend (albeit I think that has a very lot to do with next.js… and with how easy it is to make it slow for someone not that experienced or someone not being extremely careful).

Frontend is not yet as matured as JS/TS (whatever matured is, but the count of frontend frameworks is at least a magnitude higher in JS/TS), but I think when I would start a new company I would default to Rust now as frontend indeed, the language itself is for me reason. And I think vanilla-js (or Rust?) is not that much worse (time/effort-wise, sanity etc.) for more complex applications than what the Next.js ecosystem has produced so far.

Actually, my (not that small) Rust projects now take officially less time to cold compile than the “hot” reloading of our next.js monster in my job. Incremental compilation is at least an order of magnitude faster. And cherry on top, dumb code is often 100x faster than js.

You can’t imagine how often I just sweared today about js. What did go through the mind of their designers, when they created this growing disease, and why did web browsers accept this as the lingua franca for the web. So… much… pain…

Definitely not your average Rust code, more like a very ugly example of it.

Also, as the syntax first put me off as well, I gave it a chance years afterwards, and have now (or rather years ago) officially joined the church of Rust evangelism.

A lot of the syntax you define as ugly makes sense when you learn it, it’s just so much more explicit than a more dynamic language, but that exactly saves your ass a lot (it did for me at the very least) (I don’t mean macros, macros are ugly and should be avoided if possible)

Right… And the best tool for every job is of course Rust.

From Scratch (as much as I like Rust, it’s very likely more verbose from scratch). Haskell is perfect for these kinds of things.

Independently of what your position to vibe-coding or LLMs are: Vibe coding just isn’t any programming paradigm. A programming paradigm describes the structure of the program, often on a grammatical (programming language) level (e.g. declarative vs imperative). While “Vibe Coding” can lead to using one or the other paradigm, but is not a paradigm itself, it’s a tool to achieve that, similar as using an IDE with code-completion to generate code.

I’ll use it also often. But when the situation is complex and needs a lot of context/knowledge of the codebase (which at least for me is often the case) it seems to be still worse/slower than just coding it yourself (it doesn’t grasp details). Though I like how quick I can come up with quick and dirty scripts (in Rust for the Lulz and speed/power).

Ughh I tried the gemini model and I’m not too happy with the code it came up with, there’s a lot of intrinsities and concepts that the model doesn’t grasp enough IMO. That said I’ll reevaluate this continuously converting large chunks of code often works ok…

Not sure about the communities you’re visiting, the subreddits I seldom visit (because enshitification) have rather smart people.

Basically, the industry is not investing in new blood.

Yeah I think it makes sense out of an economic motivation. Often the code-quality of a junior is worse than that of an AI, and a senior has to review either, so they could just directly prompt the junior task into the AI.

The experience and skill to quickly grasp code and intention (and having a good initial idea where it should be going architecturally) is what is asked, which is obviously something that seniors are good at.

It’s kinda sad that our profession/art is slowly dying out because juniors are slowly replaced by AI.

but it can be a very helpful assistant.

can, but usually when stuff gets slightly more complex, being a fast typewriter is usually more efficient and results in better code.

I guess it really depends on the aspiration for code-quality, complexity (yes it’s good at generating boilerplate). If I don’t care about a one-time use script that is quickly written in a prompt I’ll use it.

Working on a big codebase, I don’t even get the idea to ask an AI, you just can’t feed enough context to the AI that it’s really able to generate meaningful code…

But it’s still far more political here…

Lemmy is no single instance, you can just move to a different one.

You may also want to checkout xilem

Actually that’s one of the few cases, where a (distributed/decentralized) blockchain really makes sense (trustless ledger which can be used for incorruptible/transparent political systems)…

Ignoring all the buzzword bingo and hype.

Less consumerism, more focus on real social aspects:

• Macro: robust (decentralized) political system, that’s not easily corruptible, e.g. via something like blockchain
• Micro: more focus on direct interaction with other people, not via something like a screen, as another post here already said, we’re harming ourselves (promote psychiatric issues etc.) with the current state of technology (smartphone overuse). We have gone much less social (direct interaction with others) because of this I’m sure of.

This! I feel it myself, my ADHD was much better when I stayed in a relatively natural setting with only little technology. for a few weeks (I did some programming there though, and boy was I focused in complex problems without medication etc. had one of my best coding sessions there I think). I’m pretty sure that a lot of ADHD but also other psychiatric issues like autism or social anxiety etc. that is diagnosed these days is because of all this unhealthy environment we have created. Or in other words, our modern technology promotes psychiatric issues such as ADHD, autism, social anxiety etc.