You must log in or # to comment.
Eh. I never considered myself some hard-core old professional, but:
The LLM will not interact with the developers of a library or tool, nor submit usable bug reports, or be aware of any potential issues no matter how well-documented
If an LLM introduces a dependency, I will sure as hell go see it myself. Enough people do not do that for this to become a problem?
ever heard of node.js?
Heard, not used though. Jokes about isEven™ too, but I never thought it goes like this in anything intended for external use
there’s at least one guy i know of on github whose claim to fame is he finds code in existing node codebases by big corpos that’s duplicated, breaks it out into a library, then PRs the original codebase with “instead of doing <this part> manually, switch to depending on this library”, then adds to his profile “my code is used by <big corpo>”. he had thousands of libraries like that last i checked, most of them less than ten lines of code. the manifest and other boilerplate is way larger than the actual code.
Damn. isEven come alive. But hilarious enough to watch someone do it :)
Your
node_modulesdirectory can get so bloated that the community came up with different package managers just for deduplication!pnpm, for example, makes one global-adjacent cache, and then just symlinks the dependencies as needed. This is because the regularnpmdoesn’t, because what if the package changed between the 20ms since I downloaded it fornuxt? (Sorry Nuxt users, had to pick a name)Given an example from another reply… yeah. Things are fucked
Vibe coding is a black hole. I’ve had some colleagues try and pass stuff off.
What I’m learning about what matters is that the code itself is secondary to the understanding you develop by creating the code. You don’t create the code? You don’t develop the understanding. Without the understanding, there is nothing.
I smell clickbait
Interesting. I thought this will be another post about slop PRs and bug reports but no, it’s about open source project not being promoted by AI and missing on adoption and revenue opportunities.
So I think we definitely see (and will see more) ‘templatization’ of software development. Some ways of writing apps that are easy to understand for AI and are promoted by it will see wider and wider adoption. Not just tools and libraries but also folder structures, design patterns and so on. I’m not sure how bad this will be long term. Maybe it will just stabilize tooling? Do we really need new React state management library every 6 months?
Hard to tell how will this affect the development of proper tools (not vibe coded ones). Commercial tools struggling to get traction will definitely suffer but most of the libraries I use are hobby projects. I still see good tools with good documentation getting enough attention to grow, even fairly obscure ones. Then again, those tools often struggle with getting enough contributors… Are we going to see a split between vibe coded template apps for junior devs and proper tools for professionals? Will EU step in and found the core projects? I still see a way forward so I’m fairly optimistic but it’s really hard to predict what will happen in a couple of years.
LLMs definitely kills the trust in open source software, because now everything can be a vibe-coded mess and it’s sometimes hard to check.
yeah it’s to the point now where if I see emojis in the readme.md on the repo I just don’t even bother.
I used to use emojis in my documentation very lightly because I thought they were a good way to provide visual cues. But now with all the people vibe coding their own readme docs with freaking emojis everywhere I have to stop using them.
Mildly annoying.
Check out this one I came across earlier - https://github.com/Jtensetti/fediverse-career-nexus/blob/main/README.md
It’s a federated LinkedIn. ofc it’s vibe coded.
Man… of all the vibe coding tools, Lovable has gotta be one of the most useless, too.
I work with people (all middle managers) who love Loveable because they can type a two sentence description of an app and it will immediately vomit something into existence. But the code it generates is an absolute disaster and the UIs it designs (which is supposed to be its main draw) is some of the most generic crap I’ve ever seen.
0/10, do not recommend.
well to be fair you don’t even need to look at the md since right at the top it says it’s built with loveable.
Who actually tried this?
A handful of people - https://mastodon.social/tags/noltosocial
I just wanna say that’s such a good thumbnail
Oh yeah. If it was drawn by AI, well, it sure fooled me.
A Matrix guard thing but with cat details?
Btw, how do typesetters call that kind of image? I’ve seen “hero-image” in some newspapers’ html/css.
If the abominable intelligence is killing every corner of things we consider good its time to start killing the “AI”…
Open source is not only about publishing code: it’s about quality, verifiable, reproducible code at work. If LLMs can’t do that, those “vibe coding” projects will hit a hard wall. Still, it’s quite clear they badly impact the FOSS ecosystem.
This isn’t the problem with the AI, it’s the problem with the user. If you don’t know enough to select the library and make the AI use it, maybe you were never gonna finish the project without AI anyway.
