Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cursor told me I should learn coding instead of asking it to generate it

forum.cursor.com

Post reply on HN
Print view View original

331–340 of 411 posts

Re: Cursor told me I should learn coding instead of asking it to generate it

#331

Post by kyleee »

datadrivenangel wrote: post #323
jeandesuis wrote: post #164

Sheesh, I didn't expect my post to go viral. Little explanation: I downloaded and run Cursor for the first time when this "error" happened. Turned out I was supposed to use agent instead of inline Cmd+K command because inline has some limitations while agent not so much. Nevertheless, I was surprised that AI could actually say something like that so just in case I screenshotted it - some might think it's fake, but it…

I also had a fun cursor bug where the inline generation got stuck in a loop and generated a repeating list of markdown bulletpoints for several hundred lines until it decided to give it a break.

I’ve seen similar sort of nonsense suggested by GitHub copilot on occasion

Re: Cursor told me I should learn coding instead of asking it to generate it

#332

Post by owenpalmer »

ahofmann wrote: post #233
gazereth wrote: post #188

Earlier quoted context omitted.

Personally I've found that you need to define the strategy yourself, or in a separate prompt, and then use a chain-of-thought approach to get to a good solution. Using the example you gave: Hey Chat, Write me some basic rust code to download a url. I'd like to pass the url as an string argument to the file Then test it and expand: Hey Chat, I'd like to pass a list of urls to this script and fetch them one by one. Can…

I'm going the exact opposite way. I provide all important details in the prompt and when I see that the LLM understood something wrong, I start over and add the needed information to the prompt. So the LLM either gets it on the first prompt, or I write the code myself. When I get the "Yes, you are right ..." or "now I see..." crap, I throw everything away, because I know that the LLM will only find shit "solutions".

This is actually a great approach. Essentially you're using time travel to prevent misunderstandings, which prevents the context from getting clogged up with garbage.

Re: Cursor told me I should learn coding instead of asking it to generate it

#333

Post by owenpalmer »

hakaneskici wrote: post #258
gazereth wrote: post #188

Earlier quoted context omitted.

Personally I've found that you need to define the strategy yourself, or in a separate prompt, and then use a chain-of-thought approach to get to a good solution. Using the example you gave: Hey Chat, Write me some basic rust code to download a url. I'd like to pass the url as an string argument to the file Then test it and expand: Hey Chat, I'd like to pass a list of urls to this script and fetch them one by one. Can…

I have heard a few times that "being nice" to LLMs sometimes improves their output quality. I find this hard to believe, but happy to hear your experience. Examples include things like, referring to LLM nicely ("my dear"), saying "please" and asking nicely, or thanking. Do these actually work?

I'm not sure encouragment itself is the performance enhancer, it's more that you're communicating that the model has the right "vibe" of what your end goal is.

Re: Cursor told me I should learn coding instead of asking it to generate it

#334

Post by namaria »

sshine wrote: post #328
enugu wrote: post #268

Earlier quoted context omitted.

The Turing test is meant to see if computers can functionally replicate human thought (just focussing on textual conversations aspect for simplicity). The implications of passing the Turing Test are profound. Tricking a user with a 5 minute conversation isn't the same. The examiner is allowed to be an expert questionner. For one thing, software engineering would quickly be taken over as a client can just chat with an…

It's been a while since I read the original article, https://courses.cs.umbc.edu/471/papers/turing.pdf > The Imitation Game > "Can machines think?" We may need to redefine what these words mean. Whether machines can think, and whether we believe so, are distinct. If we're just a kind of machine, we can just say that LLMs have not surpassed us. While LLMs still give away obvious signatures (their flaws are LLM-y and n…

> If we're just a kind of machine

That is not true. Even our cells, with features that look a lot like machinery such as a proton pump, is on the whole several orders of magnitude more complex than any machine. Even a single human cell is more like an ecosystem than a machine. Let alone entire humans or even just the human brain. Consider that both cells and humans are capable of reproduction.

> If anything, the ability of LLMs to imitate human labor

Within extremely narrow confines and quite often going over into just conjuring up nonsense.

Re: Cursor told me I should learn coding instead of asking it to generate it

#335

Post by croes »

aleph_minus_one wrote: post #312
croes wrote: post #280

Earlier quoted context omitted.

SO and Google are still a problem especially they are one of the sources of LLMs. So now we can get wrong code but written in a more confident language. If you are lucky it’s a hallucination and the error is obvious.

> So now we can get wrong code but written in a more confident language. The lesson to learn is rather that also in "real life", you shouldn't trust confident people.

Mean’s don’t trust AI

Re: Cursor told me I should learn coding instead of asking it to generate it

#336

Post by namaria »

Nathanba wrote: post #211
alonsonic wrote: post #124

Earlier quoted context omitted.

It's ironic to see people say this type of things and not think about old software engineer practices that are now obsolete because overtime we have created more and more tools to simplify the craft. This is yet another step in that evolution. We are no longer using punch cards or writing assembly code, and we might not write actual code in the future anymore and just instruct ais to achieve goals. This is progress

well the only issue I have with that is that coding is already a fairly easy way to encode logic. Sure...writing Rust or C isn't easy but writing some memory managed code is so easy that I wonder whether we are helping ourselves by removing that much thinking from our lives. It's not quite the same optimization as building a machine so that we don't have to carry heavy stones ourselves. Now we are building a machine…

Management has resented people capable of creating non trivial, novel and valuable software for several decades and trying to get LLMs to generate code for them is just the latest chapter in this power struggle.

Re: Cursor told me I should learn coding instead of asking it to generate it

#337

Post by daryl_martis »

IshKebab wrote: post #209
jeandesuis wrote: post #164

Sheesh, I didn't expect my post to go viral. Little explanation: I downloaded and run Cursor for the first time when this "error" happened. Turned out I was supposed to use agent instead of inline Cmd+K command because inline has some limitations while agent not so much. Nevertheless, I was surprised that AI could actually say something like that so just in case I screenshotted it - some might think it's fake, but it…

It's probably learnt it from all the "homework" questions on StackOverflow.

"learnt"

Re: Cursor told me I should learn coding instead of asking it to generate it

#338

Post by sebmellen »

ahofmann wrote: post #233
gazereth wrote: post #188

Earlier quoted context omitted.

Personally I've found that you need to define the strategy yourself, or in a separate prompt, and then use a chain-of-thought approach to get to a good solution. Using the example you gave: Hey Chat, Write me some basic rust code to download a url. I'd like to pass the url as an string argument to the file Then test it and expand: Hey Chat, I'd like to pass a list of urls to this script and fetch them one by one. Can…

I'm going the exact opposite way. I provide all important details in the prompt and when I see that the LLM understood something wrong, I start over and add the needed information to the prompt. So the LLM either gets it on the first prompt, or I write the code myself. When I get the "Yes, you are right ..." or "now I see..." crap, I throw everything away, because I know that the LLM will only find shit "solutions".

This is the best approach and avoids long context windows that get the LLM confused

Re: Cursor told me I should learn coding instead of asking it to generate it

#339

Post by namaria »

rkuodys wrote: post #231
andai wrote: post #102

This isn’t just about individual laziness—it’s a systemic arms race towards intellectual decay.[0] With programming, the same basic tension exists as with the more effective smarter AI-enhanced approaches to conceptual learning: effectiveness is a function of effort, and the whole reason for the "AI epidemic" is that people are avoiding effort like the plague. So the problem seems to boil down to, how can we convince…

For me the question is different- you can ask "how to become better coder" for which the answer might be - to write more code. But the question I'm asking is what's the job of system/software developer. And to that the answer is very different. At least for me it is"building solutions to business problems, in the most efficient way". So in that case more/better code does not necessarily bring you closer to your goal.

Efficient exists on multiple dimensions. Generating software quickly and cost effectively is just one of them. Maintaining, adding features and managing security liabilities also factor into that overall efficiency. And the ability to managing this maintenance, growth and security concerns depends on the continued ability to reason about any facet of the software that might become relevant - something that programmers writing code have shown to be capable of, but LLMs generating text continuations that contain code have not.

Re: Cursor told me I should learn coding instead of asking it to generate it

#340

Post by enugu »

sshine wrote: post #328
enugu wrote: post #268

Earlier quoted context omitted.

The Turing test is meant to see if computers can functionally replicate human thought (just focussing on textual conversations aspect for simplicity). The implications of passing the Turing Test are profound. Tricking a user with a 5 minute conversation isn't the same. The examiner is allowed to be an expert questionner. For one thing, software engineering would quickly be taken over as a client can just chat with an…

It's been a while since I read the original article, https://courses.cs.umbc.edu/471/papers/turing.pdf > The Imitation Game > "Can machines think?" We may need to redefine what these words mean. Whether machines can think, and whether we believe so, are distinct. If we're just a kind of machine, we can just say that LLMs have not surpassed us. While LLMs still give away obvious signatures (their flaws are LLM-y and n…

Yes, the Turing Test tries to give an operational definition of 'think'- can a computer give the same response as a thinking human.

Further, we can even disallow decisions based on the human/machine signature. The computers don't need to imitate human frailty and examiners decide based on the quality of the output, not whether it feels human or machine. I think that this is also why one of the stipulations is that the human/machine is not physically visible.

But even then, this goal is far away. We are not asking just about simple short chat imitation games but whether a computer can operationally do everything humans are capable of in the textual medium. This often requires experts.

The chat game version can be solved. But, the foundational goal is far away.

Post reply on HN