Interestingly enough, https://fastpotify.rocks/winamp/ shows what I dislike about LLM-generated text: From "The buttons do what they say [...]" to "[...] a different format and are not supported" is just a huge wall of text without any visual guidance.

It's as if the person writing these lines didn't care about the result. One could argue that people not caring about written documentation isn't anything new, but in the past, these documentations also had little text. So they were easier to digest. Now, we get great walls of text that are hard to read. I cannot say that I like that.

▲

embedding-shape a day ago | parent | next [-]

What sucks is that clearly the author has prompted the model somehow to generate this text, and this prompt probably contains the actual meat that the model just "expanded on" while adding it's own fluff, and if the author just shared the actual verbatim prompt instead, the documentation would be 90% shorter but also 200% more "informative per character used".

Just share the prompt you used, regardless of how low quality or bad it is, rather than this word spaghetti of 100s of words saying basically nothing, such a waste.

▲

sceptic123 21 hours ago | parent | next [-]

That isn't going to be very helpful is it?

> can you build a website to promote the app

> can you add a page showing off the winamp mode

	▲	jeltz 19 hours ago \| parent [-]
		Still more useful than what exists now.

▲

km144 17 hours ago | parent | prev [-]

That would actually be a reasonable thing to do with a model that is good at writing though. I'm sure if you used that method on Opus 4.6 you'd get a pretty decent result, because Claude models used to be quite good at that sort of technical synthesis. Asking a model that is bad at technical writing (e.g. Opus/Fable 5) to write your docs is a bad idea.

	▲	embedding-shape 16 hours ago \| parent [-]
		Ultimately what seems more important that how good the model is at task X vs task Y, seems to be that the human accepting/rejecting/iterating on the text can actually identify "good" from "bad" text, and makes the effort to fix it. Otherwise it doesn't matter much what model you use, it makes boring, bland and bloated prose.

▲

moritonal a day ago | parent | prev | next [-]

It's a weird world where they put less time into writing that doc than you did reading it.

	▲	unshavedyak 16 hours ago \| parent [-]
		The sad thing is that may not be true. If you use LLMs to write docs and actually read it, you can spend far longer trying to convince the LLM to write in a sane way than to just write it yourself. Semi related, but i've caught myself nitpicking small things from the code of an LLM only to realize i'm putting in more work trying to convince the LLM to make the code the way i want rather than just modifying it myself lol. It's as if i'm trying to educate it in the ideal way, as if it would learn and not repeat the mistake.. but it rarely does of course. It seems to me that interacting with LLMs has a large pile of cognitive traps for humans.

▲

earcar 21 hours ago | parent | prev | next [-]

I prioritized shipping over carefully crafting docs. I thought nobody would read them and they'd rather enjoy the software. I was wrong. I'll rewrite them.

	▲	tepmoc 19 hours ago \| parent \| next [-]
		Then, just do not add text that adds no value. Less is more. Documentation is first part people see when presented. I personally hate AI written text even though I do agentic engineering a lot as well. Any user facing documentation must be reviewed, if you snoozy by reading it - remove or rewrite. I do not care about paragraph long comments in code, because I’m not gonna read them, but I still follow my internal guideline and cut all this verbosity to level where it readable using hooks and skills. So basicly like you said if you care about product do not add placeholders. It’s take away first impression.
	▲	MrGilbert 20 hours ago \| parent \| prev \| next [-]
		Fair enough. But then, how should I enjoy the software, if the documentation to its' usage is buried beyond a whole paragraph? To be clear, I do not judge the usage of AI per se. But you can instruct AI to write a clear manual. It just takes time. :)
	▲	nuancebydefault 21 hours ago \| parent \| prev [-]
		I think you got a but too much negativity, from a crowd that became more easily irratated because their AI sensor became very sensitive by e.g. reviewing a lot of AI generated code and comments.

▲

alasano a day ago | parent | prev | next [-]

Not running code from someone who has so little self awareness that they publish documentation like this tbh.

Doesn't matter if it's open source.

▲

bityard 21 hours ago | parent | prev [-]

I never thought ubiquitous human-written marketing wank was any better.

	▲	t-3 20 hours ago \| parent [-]
		The difference is that the marketing stuff used to be only on corporate websites. Now it's everywhere, because pretty much everybody hates writing, and the "pretty much" can be left off when the writing is a job or some other form of obligation.