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

My article on why AI is great (or terrible) or how to use it

matthewrocklin.com

Post reply on HN
Print view View original

181–190 of 241 posts

Re: My article on why AI is great (or terrible) or how to use it

#181

Post by falloutx »

danielbln wrote: post #171
falloutx wrote: post #73

Earlier quoted context omitted.

Its sad we have gone from making the software as good and efficient as possible to "good enough, ship it". This is the main reason AI agents are successful.

Let's not rewrite history. Somehow the counter point to these AI discussions seems to be some revisionist fantasy past where humans would only write perfect code at the height of the craft. That's just not true, there was plenty of shit code around before and business interests never cared one iota about "clean" code. So no, we haven't gone from "making software as good and efficient as possible", that was always a n…

I meant not just recently, not just because of AI but there was a time when we were required to write efficient code because of hardware limitations and inability to patch quickly. AI is still not a having a massive impact on software dev yet, but it benefits from last 2 decades of lowering standards.

Re: My article on why AI is great (or terrible) or how to use it

#182

Post by saxenaabhi »

mcintyre1994 wrote: post #177

I disagree with them on the large PR section. I agree that we sometimes don’t review these in detail, but the thing that enables that for me is a trusted contributor explaining deterministically what they did. I renamed whatever variable, or I changed whatever script in all these places, I updated whatever lint rule and fixed all the errors, I moved whatever to a separate package. All of these can create a diff acros…

Ah, Human beings! Famous for their determinism, reliability, and for not making logical errors!

Re: My article on why AI is great (or terrible) or how to use it

#184

Post by mcintyre1994 »

saxenaabhi wrote: post #182
mcintyre1994 wrote: post #177

I disagree with them on the large PR section. I agree that we sometimes don’t review these in detail, but the thing that enables that for me is a trusted contributor explaining deterministically what they did. I renamed whatever variable, or I changed whatever script in all these places, I updated whatever lint rule and fixed all the errors, I moved whatever to a separate package. All of these can create a diff acros…

Ah, Human beings! Famous for their determinism, reliability, and for not making logical errors!

There are some that I trust more than I trust any LLM!

Re: My article on why AI is great (or terrible) or how to use it

#185

Post by pona-a »

tverbeure wrote: post #132
emseetech wrote: post #15

Earlier quoted context omitted.

My question is why use AI to output javascript or python? Why not output everything in C and ASM for 500x performance? Why use high level languages meant to be easier for humans? Why not go right to the metal? If anyone's ever tried this, it's clear why: AI is terrible at C and ASM. But that cuts into what AI is at its core: It's not actual programming, it's mechanical reproduction. Which means its incapabilities in…

The premise of your question is wrong. I would still write Python for most of my tasks even if I were just as fast at writing C or ASM. Because the conciseness and readability of the code that I use is way more important than execution speed 99% of the time. I assume that people who use AI tools still want to be able to make manual changes. There are hardly any all or nothing paradigms in the tech world, why do you a…

The promise of the original definition of vibe coding was you treating code as disposable, no more valuable than LLVM build cache.

You aren't supposed to make corrections, review it, or whatever.

Re: My article on why AI is great (or terrible) or how to use it

#186

Post by tonfa »

pona-a wrote: post #185
tverbeure wrote: post #132

Earlier quoted context omitted.

The premise of your question is wrong. I would still write Python for most of my tasks even if I were just as fast at writing C or ASM. Because the conciseness and readability of the code that I use is way more important than execution speed 99% of the time. I assume that people who use AI tools still want to be able to make manual changes. There are hardly any all or nothing paradigms in the tech world, why do you a…

The promise of the original definition of vibe coding was you treating code as disposable, no more valuable than LLVM build cache. You aren't supposed to make corrections, review it, or whatever.

But the LLM does, and some amount of conciseness/readability will help.

Re: My article on why AI is great (or terrible) or how to use it

#187

Post by dist-epoch »

hmry wrote: post #178
AdieuToLogic wrote: post #145

The author presents a false dichotomy when discussing "Why Not AI". ... there are some serious costs and reasonable reservations to AI development. Let's start by listing those concerns These are super-valid concerns. They're also concerns that I suspect came around when we developed compilers and people stopped writing assembly by hand, instead trusting programs like gcc ... Compilers are deterministic , making thei…

If LLMs were like compilers, you could put src/ into .gitignore and only upload the prompt. Even the earliest compilers didn't work by the programmer writing code, copying the assembly output into their source tree, and throwing away the code. This is not a value judgement, they simply aren't the same thing at all.

here you go, a prompt only library: https://github.com/dbreunig/whenwords

Re: My article on why AI is great (or terrible) or how to use it

#188

Post by dist-epoch »

sodapopcan wrote: post #12
deergomoo wrote: post #2

I will never as long as I live understand the argument that AI development is more fun. If you want to argue that you’re more capable or whatever, fine. I disagree but I don’t have any data to disprove you. But saying that AI development is more fun because you don’t have to “wrestle the computer” is, to me, the same as saying you’re really into painting but you’re not really into the brush aspect so you pay someone…

Seems some people I know who really like AI aren't particularly good with their editors. Lots of AI zealots use the "learn your tools" when they are very slow with their editors. I'm sure that's not true across the board, but the sentiment that it's not worth it to get really advanced with your editor has been pretty prevalent for a very long time. I don't care if you use AI but leave me alone. I'm plenty fast withou…

You can write 10000 lines of code without AI in a day? Impressive

Re: My article on why AI is great (or terrible) or how to use it

#189

Post by bdauvergne »

Seems people read the blog but not the code, I looked at the stated rewrite of Numpy in Rust:

> As an introductory project, I rewrote Numpy in Rust. It was great fun.

That's not a rewrite at all it's just a wrapping of an existing linear algebra Rust library (faer, blas, etc..) with a more Numpy like API. It seems to me that every AI project I look at is just a mashup/wrapper over existing things. Where are the real bootstrapped new things with AI ? Is there any big OSS project (Linux kernel, postgresql, Django, whatever) with serious bugfixes or new features implemented by AI that we could look at ?

Are so much people in programming implementing middleware / wrapping existing API all day that it gives them a feeling of liberation to be able to delegate those tasks ?

Re: My article on why AI is great (or terrible) or how to use it

#190

Post by nl »

AdieuToLogic wrote: post #162
nl wrote: post #152

Earlier quoted context omitted.

> Compilers are deterministic, making their generated assembly code verifiable People keep saying this like it is an absolute fact, whereas in reality it is a scale. Compilers are more deterministic than LLMs in general, but no they are not completely deterministic. That's why making reproducible builds is hard! https://stackoverflow.com/questions/52974259/what-are-some-e... and https://github.com/mgrang/non-determin…

>> Compilers are deterministic, making their generated assembly code verifiable > People keep saying this like it is an absolute fact, whereas in reality it is a scale. My statement is of course a generalization due to its terseness and focuses on the expectation of repeatable results given constant input, excluding pathological definitions of nondeterminism such as compiler-defined macro values or implementation def…

> Assuming the LLM in use has, or is reasonably expected to have, model evolution, documents generated by same will diverge unpredictably given a constant prompt.

So what?

You tell it once. It writes code.

You test that code, not the prompt.

Post reply on HN