Hacker News new | ask | show | jobs
by esseph 338 days ago
Some of us actually write like that... :/

Sucks when everything you write sounds like a bot because you're autistic.

3 comments

> Some of us actually write like that

The fact that LLMs write like that is proof of that people write like this too, as LLMS produce statistical averages of the input writings.

It kind of surprises me that they never confuse "it's" and "its" and common mistakes like that, when it seems like most human writers today swap them randomly. I suppose that's thanks to a lot of the text in the training data predating the collapse of English education.

I'm not sure why em dashes are so popular, though. I don't think I've ever seen human writing that had as many em dashes as LLMs use.

I add them where it makes sense, in more of a way to mimic natural conversations.
I don't think it should suck at all. As I see it, RLHF empirically validated that this is the writing style people find to be the best.
Maybe you've never been told you sound like a robot, then.

Feeling less-than-human isn't great.

Are you the post author?
No, just a human getting tired of things I type online being thought of as LLM spam.
This sounds like what a bot programmed to engage in HN would do, honestly.
I thought I would have more opinions from ivory, gated towers tbh, and maybe a fatter bank account.
I doubt it to be honest. It's so distinct and non-existing.

>With Astro you're not locked into a single way of doing things. Need React for a complex form? Chuck it in. Prefer Vue for data visualisation? Go for it. Want to keep most things as simple Astro components? Perfect.

>What struck me most after migrating several projects is how Astro makes the right thing the easy thing. Want a fast site? That's the default. Want to add interactivity? Easy, but only where you need it. Want to use your favourite framework? Go ahead, Astro won't judge.

>Developer experience that actually delivers

I am downvoted so I guess I'm wrong. It's just bland and form in a way ChatGPT usually outputs. Sorry to the author if I'm wrong.

I've never seen an LLM use this sort of socratic writing style.
Damn, first it sounds like a robot, then it's bland.

Speak your truth, poster!