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

I'd rather read the prompt

claytonwramsey.com

Post reply on HN
Print view View original

131–140 of 880 posts

Re: I'd rather read the prompt

#131

Post by jez »

Where I especially hold this viewpoint is for end-of-year peer performance reviews.

People say “I saved so much time on perf this year with the aid of ChatGPT,” but ChatGPT doesn’t know anything about your working relationship with your coworker… everything interesting is contained in the prompt. If you’re brain dumping bullet points into an LLM prompt, just make those bullets your feedback and be done with it? Then it’ll be clear what the kernel of feedback is and what’s useless fluff.

Re: I'd rather read the prompt

#132

Post by mopenstein »

xmorse wrote: post #20

I am thinking about creating a proof-of-writing signature. Basically an editor with an "anti-cheat", you can't paste text into it. It signs your text with a public key.

Can't a raspberry pi (or similar) emulate a USB keyboard? Feed it any text and the key strokes will look real to your editor.

I guess you could require a special encrypted keyboard in your plan.

Re: I'd rather read the prompt

#133

Post by generativenoise »

throwawaysleep wrote: post #125
jsheard wrote: post #74

I'm far from the first to make this observation but LLMs are like anti-compression algorithms when used like that, a simple idea gets expanded into a bloated mess by an LLM, then sent to someone else who runs it through another LLM to summarize it back to something approximating the original prompt. Nobody benefits aside from Sam Altman and co, who get to pocket a cool $0.000000001 for enabling this pointless exercis…

Depends on what you are looking for. I’ve turned half baked ideas into white papers for plenty of praise. I’ve used them to make my Jira tickets seem complicated and complete. I’ve used them to get praised for writing comprehensive documentation. Part of my performance review is indirectly using bloat to seem sophisticated and thorough.

Would be nice to fix the performance reviews so we don't end up in a arms race of creating bloat until it becomes so unproductive it kills the host.

Over-fitting proxy measures is one of the scourges of modernity.

The only silver lining is if it becomes so wide spread and easy it loses the value of seeming sophisticated and thorough.

Re: I'd rather read the prompt

#134

Post by staunton »

otabdeveloper4 wrote: post #88
mrweasel wrote: post #19

Earlier quoted context omitted.

> Most people do view obtaining a degree as just a hurdle to getting a decent job Then fail to actually learn anything and apply for jobs and try to cheat the interviewers using the same AI that helped them graduate. I fear that LLMs have already fostered the first batch of developers who cannot function without it. I don't even mind that you use an LLM for parts of your job, but you need to be able to function witho…

Lots and lots of developers can't program at all. As in literally - can't write a simple function like "fizzbuzz" even if you let them use reference documentation. Many don't even know what a "function" even is. (Yes, these are people with developer jobs, often at "serious" companies.)

I've never met someone like that and don't believe the claim.

Maybe you mean people who are bad at interviews? Or people whose job isn't actually programming? Or maybe "lots" means "at least one"? Or maybe they can strictly speaking do fizzbuzz, but are "in any case bad programmers"? If your claim is true, what do these people do all day (or, let's say, did before LLMs were a thing...)?

Re: I'd rather read the prompt

#135

Post by necovek »

tptacek wrote: post #124
necovek wrote: post #112

Earlier quoted context omitted.

I think they summed it up well in the section "Why do we write, anyway?" — they nowhere claimed it was vital for students' success in a robotic class. On the contrary as they title a subsection there with "If it’s worth doing, it’s worth doing *badly*" (emphasis mine) — so what they are looking for is to peer into the author's mind and their original thoughts. The implication there is that this is acceptable to pass…

Who have you asked? What has been your experience observing professionals using LLMs?

Is "what has been my experience" not implied in what I am still waiting for — "someone using an LLM to produce high quality code in the field of their expertise in less time than without an LLM"?

So, observing a couple of my colleagues (I am an engineering manager, but have switched back and forth between management and IC roles for the last ~20 years), I've seen them either produce crap, or spend so much time tuning the prompts that it would have been faster to do it without an LLM. They mostly used Github Copilot or ChatGPT (most recent versions as of last few months ago).

I am also keeping out a keen eye for any examples of this (on HN in particular), but it usually turns out things like https://news.ycombinator.com/item?id=43573755

Again, I am not saying it's not being done, but I have struggled to find someone who would demonstrate it happen in a convincing enough fashion — I am really trying to imagine how I would best incorporate this into my daily non-work programming activities, so I'd love to see a few examples of someone using it effectively.

Re: I'd rather read the prompt

#136

Post by johnfn »

vunderba wrote: post #31
bost-ty wrote: post #4

I like the author's take: it isn't a value judgement on the individual using ChatGPT (or Gemini or whichever LLM you like this week), it's that the thought that went into making the prompt is, inevitably, more interesting/original/human than the output the LLM generates afterwards. In my experiments with LLMs for writing code, I find that the code is objectively garbage if my prompt is garbage. If I don't know what I…

This is the CRUX of the issue. Even with SOTA models (Sonnet 3.5, etc) - the more open-ended your prompt - the more banal and generic the response. It's GIGO turtles all the way down. I pointed this out a few weeks ago with respect to why the current state of LLMs will never make great campaign creators in Dungeons and Dragons. We as humans don't need to be "constrained" - ask any competent writer to sit quietly and…

> I guarantee 99% of the returns will return a very formulaic physics-based puzzle response like "The Resonant Hourglass"

Haha, I was suspicious, so I tried this, and I indeed got an hourglass themed puzzle! Though it wasn't physics-based - characters were supposed to share memories to evoke emotions, and different emotions would ring different bells, and then you were supposed to evoke a certain type of story. Honestly, I don't know what the hourglass had to do with it.

Re: I'd rather read the prompt

#137

Post by necovek »

nathants wrote: post #110
necovek wrote: post #95

Earlier quoted context omitted.

Unfortunately, the video is a far cry from carrying all the representative information: there is no way you can capture your full emotions as you are working through a problem, and where did you get your "eureka" moments unless you are particularly good at verbalising your through process as you go through multiple dead-ends and recognize how they lead you in the right direction. And reviewing video would be a nightm…

there are only two options: - have more information - have less information more is better. you can scrub video with your finger on an iphone. serious review is always high effort, video changes nothing.

Not really: I love reading fiction where I can imagine characters the way I want to based on their written depictions. When I see a book cover replaced with a recent movie adaptation actor, it usually reduces the creative space for the reader instead of enlarging it.

Video in itself is not more information by definition. Just look at those automatically generated videos when you try finding a review on an unusual product.

Re: I'd rather read the prompt

#138

Post by HPsquared »

Nezteb wrote: post #67
vunderba wrote: post #31

Earlier quoted context omitted.

This is the CRUX of the issue. Even with SOTA models (Sonnet 3.5, etc) - the more open-ended your prompt - the more banal and generic the response. It's GIGO turtles all the way down. I pointed this out a few weeks ago with respect to why the current state of LLMs will never make great campaign creators in Dungeons and Dragons. We as humans don't need to be "constrained" - ask any competent writer to sit quietly and…

Out of curiosity, I used your prompt but added "Do not make it a very formulaic physics-based puzzle." The output is pretty non-sensical: https://pastebin.com/raw/hetAvjSG

It is totally different from anything that exists. It fulfils the prompt, I suppose! It has to be crazy so you can be more certain it's unique. The prompt didn't say anything about it being good.

Re: I'd rather read the prompt

#139

Post by _qswe »

satisfice wrote: post #111
_qswe wrote: post #78

Earlier quoted context omitted.

I think the author has a fair take on the types of LLM output he has experience with, but may be overgeneralizing his conclusion. As shown by his example, he seems to be narrowly focusing on the use case of giving the AI some small snippet of text and asking it to stretch that into something less information-dense — like the stereotypical "write a response to this email that says X", and sending that output instead o…

If you present your AI-powered work to me, and I suspect you employed AI to do any of the heavy lifting, I will automatically discount any role you claim to have had in that work. Fairly or unfairly, people (including you) will inexorably come to see anything done with AI as ONLY done with AI, and automatically assume that anyone could have done it. In such a world, someone could write the next Harry Potter and it wi…

To be clear, I'm not a student, nor do I disagree with academic honor codes that forbid LLM assistance. For anything that I apply AI assistance to, the extent to which I could personally "claim credit" is essentially immaterial; my goal is to get a task done at the highest quality and lowest cost possible, not to cheat on my homework. AI performs busywork that would cost me time or cost money to delegate to another human, and that makes it valuable.

I'm expanding on the author's point that the hard part is the input, not the output. Sure someone else could produce the same output as an LLM given the same input and sufficient time, but they don't have the same input. The author is saying "well then just show me the input"; my counterpoint is that the input can often be vastly longer and less organized or cohesive than the output, and thus less useful to share.

Re: I'd rather read the prompt

#140

Post by staunton »

Herring wrote: post #117
robwwilliams wrote: post #106

I ask Claude to respond like Hemingway would. It works.

I think half the population of the world wants just the facts, the other half wants long flowing beautiful content like on apple.com, and neither group knows the other exists. Of course this is the right way to do it!

I'm sure it's beside the point but...

Your benchmark for "long flowing beautiful content" is apple.com? It's competing with Hemingway?

Post reply on HN