cross-posted from: https://lemmy.dbzer0.com/post/50693956
Transcript
A post by [object Object] (@zzt@mas.to) saying: courtesy of @davidgerard@circumstances.run, Proton is now the only privacy vendor I know of that vibe codes its apps: In the single most damning thing I can say about Proton in 2025, the Proton GitHub repository has a “cursorrules” file. They’re vibe-coding their public systems. Much secure! I am once again begging anyone who will listen to get off of Proton as soon as reasonably possible, and to avoid their new (terrible) apps in any case. https://circumstances.run/@davidgerard/114961415946154957It has a reply by the author saying: in an unsurprising update for those familiar with how Proton operates, they silently rewrote their monorepo’s history to purge .cursor and hide that they were vibe coding: https://github.com/ProtonMail/WebClients/tree/2a5e2ad4db0c84f39050bf2353c944a96d38e07f
given the utter lack of communication from Proton on this, I can only guess they’ve extracted .cursor into an external repository and continue to use it out of sight of the public
Calling something "vibe coded" feels similar to calling something "woke". Its used for everything so it lost all meaning.
If I start writing a function name and the coding assistant suggests exactly, character by character, the function signature I would have typed, is that vibecoding? If I ask an agent to check out my classes and create me a json schema from that, what about that? What if I ask it to write me some database connection stuff, boilerplate that is basically directly copy pasted from docs?
The presence of a cursorrules file signifies.. Nothing. Most projects now have a prompt file to store general context, like styling etc. Most programmers use coding assistants. The use of coding assistants does not, in of itself, mean bad quality. Obviously you have to hide it because people get irrationally angry about it.
The blind AI hate does not help. LLMs can be useful. Is it stupid to have an LLM write an email so the other side can use an LLM to summarize it? Yes. Is LLM slop annoying? Yes. That does not mean that everything is bad. If you check out the Developer-sphere, people who have tried coding assistants for a couple years now, the consensus is "overhyped but definitely useful".
Example, read this comment section: https://news.ycombinator.com/item?id=44836879
I agree. Unfortunately, they’re being forced on us by the same companies which are enshittifying things as much as they can, and respond with hostility when anyone suggests that users deserve privacy, or fair compensation for their work. I can’t really blame people who respond with skepticism and distrust.
The second you allow slop generator use into your code base the quality takes a nosedive. People claim to review their slop but I have seen it often enough, all they are doing is hitting accept all without even reading half of the generated code. This isn't "just a tool" this is technical debt as a service.
Also Hackernews is ridiculous den of techbros, libertarians, AI maximalists and temporarily embarrassed founders who think they stand to gain the most from hyping up slop generators. Trusting them with anything related to AI is like asking the CEO of Microsoft if you should use Windows or Linux.
I'm going to start vibe-Woking where I use an AI to find me new and original ways to trigger whole sections of society and use that power for good (or evil, I haven't decided yet).
Yeah, impossible to tell from that if someone is trying to "prompt engineer" all the way through a problem (which sucks and I'm shocked anyone reports that it works out for them) versus hitting tab complete for a couple of lines at a time or maybe prompt your way through a pretty boilerplate little function.
If they try to prompt engineer the whole project, well it should be obvious because it almost certainly has to be utter crap.