Movim is a federated chat platform that recently implemented most of Discord’s features. It’s a little rough in the UX department, but it’s pretty much all there, and it isn’t vibe-coded like this chatto app is.
Movim has been developed since 2011 (Chatto began development 8 months ago, in comparison) and is based on XMPP, which is battle tested and proven to scale. It has a codebase that is maintainable long-term.
I personally would not trust an almost entirely vibe-coded app using tools prone to hallucination to be secure and properly implement encryption. But if others don’t have those concerns, then by all means.
The fact that it has been around since 2011 and still suffers from UI problems and other bugs really highlights the problem with your arguments. You also very clearly have no idea how software development actually works if you think that code written manually by a human is inherently more reliable. Humans make mistakes all the time that’s why we have software bugs in the first place. The way we ensure that code works is by having things like tests, specifications, and code reviews. All of these same tools work just as well with LLM generated code as they do with code written by hand.
Movim is a federated chat platform that recently implemented most of Discord’s features. It’s a little rough in the UX department, but it’s pretty much all there, and it isn’t vibe-coded like this chatto app is.
Incredible to suggest using for an objectively worse and less mature project just because it’s artisanally written.
Movim has been developed since 2011 (Chatto began development 8 months ago, in comparison) and is based on XMPP, which is battle tested and proven to scale. It has a codebase that is maintainable long-term.
I personally would not trust an almost entirely vibe-coded app using tools prone to hallucination to be secure and properly implement encryption. But if others don’t have those concerns, then by all means.
The fact that it has been around since 2011 and still suffers from UI problems and other bugs really highlights the problem with your arguments. You also very clearly have no idea how software development actually works if you think that code written manually by a human is inherently more reliable. Humans make mistakes all the time that’s why we have software bugs in the first place. The way we ensure that code works is by having things like tests, specifications, and code reviews. All of these same tools work just as well with LLM generated code as they do with code written by hand.