Reminds me of that one messed up game[0] in 17776[1] (which I highly recommend reading from the start) Specifically in that the rules for an entirely different kind of game seem to have grown out of control.

[0] https://www.sbnation.com/a/17776-football/you-know-what-neve...

[1] https://www.sbnation.com/a/17776-football/

And I don't want to use tools I don't understand at least to some degree. I always get nervous when I do something but don't know why I do that something

Depends on what level of abstraction you're comfortable with. I have no problem driving a car I didn't build.

I didnt build my car either. But I understand a bit of most of the main mechanics, like how the ABS works, how powered steering does, how an ICE works and so on.

So far, eqch and every time I used an LLM to help me with something it hallucinated non-existant functions or was incorrect in an important but non-obvious way.

Though, I guess I do treat LLM's as a last resort longshot for when other documentation is failing me.

Knowing how to use LLMs is a skill. Just winging it without any practice or exploration of how the tool fails can produce poor results.

"You're holding it wrong"

99% of an LLM's usefulness vanishes, if it behaves like an addled old man.

"What's that sonny? But you said you wanted that!"

"Wait, we did that last week? Sorry let me look at this again"

"What? What do you mean, we already did this part?!"

Wrong mental model. Addled old men can't write code 1000x faster than any human.

I'd prefer 1x "wrong stuff" than wrong stuff blasted 1000x. How is that helpful?

Further, they can't write code that fast, because you have to spend 1000x explaining it to them.

Except it's not 1000x wrong stuff, that's the point. But don't worry, the Amish are welcoming of new luddites!

Which LLMs have you tried? Claude Code seems to be decent at not hallucinating, Gemini CLI is more eager.

I don't think current LLMs take you all the way but a powerful code generator is a useful think, just assemble guardrails and keep an eye on it.

Mostly chatgpt because I see 0 value in paying for any llm, nor do I wish to gice up my data to any llm provider

Speaking as someone who doesn't really like or do LLM-assisted coding either: at least try Gemini. ChatGPT is the absolute worst you could use. I was quite shocked when I compared the two on the same tasks. Gemini gets decent initial results you can build on. ChatGPT generates 99% absolutely unusable rubbish. The difference is so extreme, it's not even a competition anymore.

I now understand why Altman announced "Code Red" at OpenAI. If their tools don't catch up drastically, and fast, they'll be one for the history books soon. Wouldn't be the first time the big, central early mover in a new market suddenly disappears, steamrolled by the later entrants.

They work better with project context and access to tools, so yeah, the web interface is not their best foot forward.

That doesn't mean the agents are amazing, but they can be useful.

A simple "how do I access x in y framework in the intended way" shouldnt require any more context.

instead of telling me about z option it keeps hallucinating something that doesnt exist and even says its in the docs when it isnt.

Literally just wasting my time

I was in the same camp until a few months ago. I now think they're valid tools, like compilers. Not in the sense that everyone compares them (compilers made asm development a minuscule niche of development).

But in the sense that even today many people don't use compilers or static analysis tools. But that world is slowly shrinking.

Same for LLMs, the non LLM world will probably shrink.

You might be able to have a long and successful career without touching them for code development. Personally I'd rather check them out since tools are just tools.

With proton, at least, you can just use your own domain and if they ever get bad, you just point the MX record to some other service, or self host, and pretty much have it taken care of.

IMO the vidya gaem awards [0] are far superior to the game awards.

[0] https://www.youtube.com/watch?v=mXMcq_LJ8ro

Maybe you can give a bit of context why you feel that way? Dropping a 2+ hour, <2000 views, 4chan video without context isn’t really the type of comment HN is looking for as far as I can tell

Or the black hole https://floor796.com/#t3r3,28,997

Or the Ninja Turtle to the left of the black hole.

Cute. There's that tripedal robot from Interstellar there.

As someone currently looking for a new job, I stopped bothering with cover letters because they didn't make the faintest of differences. After many dozens of rejections I am just burnt out about writing them.

Just FYI the 's got swallowed by the HN formatting and made the stuff inbetween italic.

  struct { char *name; char *value}

can bypass the formatting by prepending 2 spaces

> Just FYI the 's got swallowed

Ironically enough your *'s as well.

I thought fp was more avoiding mutability/reassigning values to identifiers

What you wrote makes me think more of the point free style

When hearing of assignment statements, I think of mutability. Point free makes me think of (low to) no variables, rather than no assignments

> This is the correct answer.

Where does that saying come from? I keep seeing it in a lot of different contexts but it somehow feels off to me in a way I can't really explain.

It's not "off" unless you're simply reading it literally. If you do that, then it's a verbose way of saying "I agree". But the connotations are something like "I agree, strongly, and in particular am implying (possibly just for effect) that there are objectively right and wrong answers to this question and the other answers are wrong." The main difference is the statement that there is an objective answer to what people may be treating as a subjective question.

If it helps, you can think of it as saying more about possible disagreeing opinions than about the specific opinion expressed. "This answer is right, and the people who disagree are 'objectively' wrong."

It took me some time to catch on to this. It can certainly be jarring or obnoxious, though sometimes it can be helpful to say "yo people, you're treating this like a subjective opinion, but there are objective reasons to conclude X."

I (the comment writer), agree that it's jarring and a bit obnoxious. There were three factors that led me to write it anyway, which I've mentioned here:

https://news.ycombinator.com/item?id=46209137

Edit: Rereading the comments, I agree (heheh) with you analysis. I hadn't considered saying "I agree", because I didn't feel I was expressing an opinion, but a fact, like 1+1=2. The comment stated that the mods in fact disallow those comments and provided proof, so I didn't consider it an opinion.

Heh, and rereading my comment, it comes across as more against the usage than I actually feel. It's not my personal style, and sometimes I find it annoying, but 80% of the time I think it's totally fine and expresses a nuance that would take a lot more words otherwise. Your usage here, for example, seems totally appropriate to me.

The reason why I was reticent to use it was not because I was uncomfortable asserting an absolute (the link showed clearly that mods didn't allow these comments, I don't see any controversy there), but more so that on this type of forum, the act of voting itself is the primary method of agreement. Saying "I agree" or "this is true" or "THIS!" is generally redundant and noisy.

I really like this conversation by the way. I'm actively trying to become a better writer (by doing copywork of my favourite writers), and no other forum on Earth has this sort of conversation in such an interesting, nuanced way.

Yeah, that seems like a fair way to put my feelings into words.

It's the first time I've ever commented that, and I was trying to figure out a way to omit it. I don't like that sort of phrase either, I especially hate comments that just go "This.", but they're rare on HN so I'm in good company.

Ultimately, I put it because:

- It was the most directly informative comment on the thread;

- It had been downvoted (greyed out) to the very bottom of the thread; and

- I wanted to express my support before making a fairly orthogonal comment without whiplashing everyone.

The whiplashing concern is the problem I run into most generally. It can be hard to reply to someone with a somewhat related idea without making it seem like you're contradicting them, particularly if they're being dogpiled on with downvotes or comments. I'd love to hear other ways to go about this, I'm always trying to improve my communication.