The documentation is what gets me. Why did a bot need to write the documentation?

I can stomach using an LLM as basically a really good static analyzer (begrudgingly) but having it write the docs? How do I know you know what the code does if you didn’t even write the docs?

I’ll do you one better.

Emacs Org mode, Logseq, Vi + .MD files + grep.

I’m partial to that last one. Un-ironically I would use the closed source Obsidian before this.

I guess it depends on what we mean by “using an LLM”. I’m definitely more conservative than most but for me I draw the line at permitting an LLM to author code; while not ideal I can accept using it has a fuzzer and static analysis tool.

The community will have to agree to what its pain tolerance is and enforce that position.

No OpenBSD/NetBSD?

Removed. AI is fundamentally anti open-source.

They explicitly reference the LLM generated front end code eventually being replaced when they can “hire a proper JS/CSS” dev.

So hopefully that happens in the near future!

Love the disclosure of AI in the frontend, will be keeping and eye on this one as he replaces the frontend code.

Very cool!

The code could be perfect. Doesn’t matter; it’s slop.

Every line could be immaculate, idiomatic code. It’s slop.

This could cure cancer and it would still be slop.

There is nothing you could do to make these projects anything other than slop. Absolutely nothing can be done. They are ontologically slop.

And by extension so are you.

Nope.

Super easy to not use it. Mad easy. Mind blowingly easy.

And the benefit is, much like art, I can feel good about the software I’m using. I can connect to the developer when I read their code and try and think of what they wanted to accomplish, line by line.

No such satisfaction with the slop-factory. Just joyless masturbation.

Recommend any good pods?

I saw it, I was excited, then the disappointment hit me.

One day a human will make something like this and I can use it.

Guix.

Try Guix. It’s great.