I have been thinking about just this the past two weeks. I have been trying to recall the excitement, the joy in the most mundane things, and indeed I think it has a lot to do with anticipation. FWIW LSD allowed me to experience this joy, and the curiosity kids have, or I had. :)
Yeah, if this is how people interact with claude I’m not surprised they’re having a bad time in ways that I don’t. Asking it why it did something or getting combative is a waste of time.
Realise that you are talking to some mathematics in a box. You can't get a rise out of it. It cannot feel guilt or remorse. Whatever emotional payoff one might want from "I asked for ONE button" cannot be had here. Meanwhile, not only are you being charged by the word, but every word you send that is not directly on the path to getting what you want done is just noise in the maths getting in your way.
So leave emotion at the door and make your words count. Voice your frustrations at the stress doll next to your monitor, sure, but spending tokens on them us just wasting time and money. Explicitly typing out your exasperation will not get you closer to whatever it is you are trying to accomplish. What can you type that will?
Laser focus: clearly, concisely explain what is wrong right now, and how you want it solved. Then do that again until all the moles have been whacked. If the AI is stuck in a sycophancy loop, start a new clean session. Do this frequently anyway: every turn in the same session charges for all preceding conversation again /and/ degrades the LLM's performance.
> Meanwhile, not only are you being charged by the word,
I've never explicitly paid for AI and never will. I only use the free Claude, etc. So I'm happy to waste THEIR electricity telling their machine how much of a piece of garbage it is and hoping that sentiment gets back in some form to the dumb humans making such a dumb product. It's not like they don't evaluate performance with usage data.
(Afterwards, I do also make sure to actually click the "thumbs down" or whatever equivalent button, and submit a report if it's possible.)
You can blame the people making poorly performing LLM chatbots, not me. (For that matter, the paying users actually funding their insanity have more on their conscience than I do trying to keep up by using tools available to me and reacting in a normal way to their poor performance.)
Except you are missing an important aspect: the model includes training data that responds to THIS IS FUCKING WRONG, after which highly likely follows correction patterns and working solutions.
Ie, actually, yes, shouting at the AI can (and does, there are papers on this), work.
Anyone remember the gpt-3 era where adding "this is important to my job" and "if this is wrong someone dies" to prompts made a measurable improvement in accuracy and reduced hallucinations?
Efficient LLM interaction is context management. The golden rule: Nothing irrelevant in context.
Non-exhaustive consequences of following that rule:
* Each session handles one task, or at most, a few tightly coupled tasks.
* Emotions stay out of context; no showing frustration, no saying thanks (or at least wait until you're about to end the session)
* When possible, provide relevant files (or sections of files) instead of making the model search and read many irrelevant files.
* Keep CLAUDE.md short.
* Disable irrelevant tools.
* Revert history when the model makes mistakes. Don't make it read its mistake and fix it; fork the chat before the mistake and exclusively mention the correct action.
Attention is more limited than the context limits imply; stuff at the beginning of context stays high-attention for a while, stuff right at the end is always high-attention. If you're about to ask for something the model frequently forgets (eg, style), remind it that those instructions exist ("Following the style guidelines, implement feature X.")
My best takeaway from the site was: Every time you see Claude Code / Codex doing something you don't want, ask it to add to AGENTS.md. Specially if they churn a lot doing something, and eventually find a way. I ask it to store the way it made something work. It's a constant gardening of AGENTS.md and it has really working well.
A lot of the more sane people I work with have decided that it's futile to try and have stopped working with Opus (we can't use Fable at work due to data retention policies).
Yeah that's when the site lost me too. I feel like people just tell AI "make the thing" and then get mad when it doesn't match up to their vision that they didn't specify at all.
Same, I don't get it? I think event Sonnet 5 can build reasonable-y decent software. I'm a little surprised with how many people resonate with this.
I spent 30 mins today planning a change that affected multiple modules, and then handed it off the plan to the Claude to implement, and came back to 4 PRs fully ready to merge 15 mins later.
1 year ago, this kind of work would have taken me a better portion of the day, especially given the amount of searching I would need to do to build context.
If someone is that frustrated with AI, do they not just … switch to not using AI and doing it by hand, asking simpler next-step questions and hand-coding, instead?
But it is futile to get frustrated or annoyed by an LLM. I do get frustrated, too, but I do not "yell" at it hoping that it will miraculously do what I want. GPT (again, free tier, so no wonder) got me frustrated too because I felt like it just would not listen, no matter how specific I was, so yeah I did experience what the author intended to show.
> "yell" at it hoping that it will miraculously do what I want
It's certainly not helpful in getting the LLM from A to B, but maybe you don't care? Maybe you just need to blow off some steam?
People may call that irrational, but I'm certainly not seeing an abundance of commonly-agreed-upon rationality in all the other things they (we) do, so it's in character.
But it’s *wildly* unhelpful. Would you talk to your coworker like that? “Half the site is blue” or would you laugh and go, “Uh, whoops, now both buttons are blue, just wanted the ‘Add to Cart’ button.”
Either that, or your coworker themselves would laugh and tell you about it later.
That’s also where this quiz lost me. I wouldn’t respond in either way, I’d say “all buttons look blue now. Can we make it so that just the Add to Cart button is blue? I’m okay with Add to Cart having its own class to make it easier.” or something.
FYI I actually agree, it also surprised me. Anyway everything is deterministic on the website so it's not like that prompt actually has an influence on anything.
I believe it is similar ot other crafts. You won't be a better engineer by only looking at source code. But it does help when you are analyzing a codebase to learn something applicable to your situation. So, it is not "just reading" I believe, but a subconcious act of analyzing.
Like with any advise that ought to get you expertise or ability, if you're not goal oriented, actively gathering and acting on feedback and hyper focused on improving, you will only waste time.
Oh yeah, I have been sleep deprived so much that the things that I said made no sense. I still formulated sentences but they did not have any meaning. In fact, I noticed this myself and I was like "fuck, what I just said made no sense". This happened a few times. It is a pretty interesting experience.
I had the same experience after almost 4+ days going without sleep. A friend came to check up on me after I fell asleep, I woke up and started telling him a whole story that made no sense, but I said it with such importance that for a week he was asking me to explain to him what I meant...which I don't remember exactly, but it was important.
This sometimes happens to me when I get a migraine.
A sentence can be coherent in the formulaic sense, but complete nonsense as far as words. I immediately notice that it's incorrect, but I don't have the ability to fix it at that moment.
Yes! I immediately noticed that what I said was wrong (even though the word I said still felt somehow connected to the thought I was trying to communicate). I was really worried at first when it happened because at the first time I had no idea sleep deprivation would cause this, or that I would experience it.
I had no idea migraine causes this as well, interesting.
Sauna raises HR without increasing blood pressure though (modulo some possible initial short spike at the beginning), because your blood vessels dilate.
Cannabis causes dilation of blood vessels and elevated HR, it’s the same effect. If you take a big dab you can definitely feel your blood pressure drop and muscles relax even though your heart rate has increased.
I find it interesting how many people come out of the wood work to make disparaging comments about people who use cannabis whenever the subject of cannabis use comes up on HN.
Having lived in a country where cannabis has been legal for about a decade now it's quaint seeing this kind of casual disrespect levelled by a stranger.
I wonder if this kind of mindset is still common here and people just don't vocalize it anymore, or if it's the kind of mentality that continued criminalization perpetuates.
> how many people come out of the wood work to make disparaging comments about people who use cannabis
It's actually not even just HN. I've noticed this with many people in real life (US).
It's especially strong with the subset of Boomers who never used it. They make it part of their personality to have not used it, while often fully embracing alcohol and ignoring all of the personal and societal risks it brings. Surely they pass this mentality onto their kids with a 50% success rate.
Maybe it's some kind of inferiority/superiority compensation mechanism for not being a hippy like their rival sibling was, etc.
Weed (as prescribed) gets me and many others moving and active.
There was a deliberate effort to demonize weed in the US as it's associated with mexicans, black people, and counterculture, and it's difficult to monetize if it's legal (anyone can just grow it).
I'm pretty sure the historical reasons I mentioned involved creating the software before the SQL standard came out, meaning they were treating NULL as meaning "nothing", instead of meaning "unknown" as the SQL standard requires. But that's as far as I care to go digging into a database system I will never use of my own free will.
Yeah, because criticizing the US, and/or making China look good can get you on certain lists. Maybe that is the reason?
FWIW I do not buy into the "China bad, we good" narrative either.
The US has done really fucked up things, bombed countries for freedom, there has been recent events as well. I do not even think they were ever the "good guys" they thought they are. In Hollywood movies, sure, but in reality? Nah.
> Yeah, because criticizing the US, and/or making China look good can get you on certain lists.
What lists? And what happens to people on those lists?
> Maybe that is the reason?
If you think people are signing up throwaway accounts to post replies to random comments, making no substantive or even controversial statements, I have a bridge to sell you. In the US or China. Take your pick.
> FWIW I do not buy into the "China bad, we good" narrative either.
I never presented that narrative.
> The US has done really fucked up things, bombed countries for freedom, there has been recent events as well. I do not even think they were ever the "good guys" they thought they are. In Hollywood movies, sure, but in reality? Nah.
I'd agree. But what does this have to do with an analysis of China?
My previous comment got flagged about a minute after I posted it to you. It was three paragraphs and it argued a position without any abuse in it. It is probably someone in this thread who hit flag instead of replying because they did not like what I said. Which is suspicious, and it is also the thing I was already talking about. The throwaway shill insinuation earlier was a way to avoid engaging with a disagreement, and a flag does the same job. Welp.
reply