Live data from Hacker News

AI made coding more enjoyable

weberdominik.com

91–100 of 103 posts

Re: AI made coding more enjoyable

#91
This is the kind of person who would find more joy in telling an android to play a piano than practicing scales/arpeggios etc in order to play complicated music themselves.

I legitimately enjoy scale practice when I'm playing piano. In a similar way I've always found some pleasure in writing boilerplate and refactoring.

There is joy/peace and instructive value in the "boring" parts of almost every discipline. It's perhaps more meditative and subtle, but still very much there in abundance. And primes you much better for a real flow state.

Ever since AI exploded at my day job, I haven't legitimately been in anything resembling a programming flow state at work.

Re: AI made coding more enjoyable

#92
The split here is between AI as amplifier vs. AI as replacement. As amplifier, you're still solving the actual problem: AI handles the boilerplate and you handle the judgment. As replacement, you lose the feedback loop that makes you better over time. The developers who thrive will be the ones who know which problems still require them to be in the loop. That's a skill that takes deliberate practice and inuition to develop and almost no AI tooling is designed to teach that.

Re: AI made coding more enjoyable

#93
post #90

Earlier quoted context omitted.

> For those developers who just want to build a thing to drive business value, or because they want a tool that they need, or because they think the end result will be fun to have, AI coding is great. It enables them to skip over (parts of) the tedious coding bit and get straight to the result bit. Then they aren't programmers anymore, are they? We don't call people using no-code platforms "programmers" and we wouldn…

Right not coding with AI requires a lot of skill in understanding where the AI is going wrong, so it's still coding. Someone who can't code isn't going to make a good app with AI (although a 'working' app is definitely possible.) In the future though, sure,it'll be possible to build a decent app without ever seeing or understanding the code.

> Someone who can't code isn't going to make a good app with AI (although a 'working' app is definitely possible.)

The customer/user can't tell the difference between a good working app and a poor working app.

Re: AI made coding more enjoyable

#94
post #90

Earlier quoted context omitted.

Right not coding with AI requires a lot of skill in understanding where the AI is going wrong, so it's still coding. Someone who can't code isn't going to make a good app with AI (although a 'working' app is definitely possible.) In the future though, sure,it'll be possible to build a decent app without ever seeing or understanding the code.

> Someone who can't code isn't going to make a good app with AI (although a 'working' app is definitely possible.) The customer/user can't tell the difference between a good working app and a poor working app.

> The customer/user can't tell the difference between a good working app and a poor working app.

Come on man, this is the whole reason Duolingo was people's favourite language learning app, or people claim they like iPhones over Android phones or Photoshop over Canva. These apps and devices all work, but which one is good or better is a debate. People have preferences; some apps in a category are easier to use than others, and some apps have branding that signals status. Now, those things become more important in differentiating your app than "It does what it is supposed to do". Until now, just getting an app to do what you wanted it to do was a competitive advantage, that's becoming a smaller advantage day by day

Re: AI made coding more enjoyable

#95
post #47

Earlier quoted context omitted.

How much of "good engineering practices" exist because we're trying to make it easy for humans to work with the code? Pick your favorite GoF design pattern. Is that they best way to do it for the computer or the best way to do it for the developer? I'm just making this up now, maybe it's not the greatest example; but, let's consider the "visitor" pattern. There's some framework that does a big loop and calls the visi…

LLMs have limited working memory, like humans, and most of the practices that increase human programming effectiveness increase LLM effectiveness too. In fact more so, because LLMs are goldfish that retain no mental model between runs, so the docs had better be good, abstractions tight, and coding practices consistent such that code makes sense locally and globally.

So are we basically saying that LLMs work most effectively on codebases that exhibit good quality coding practices, but are not themselves particularly good at creating such quality code themselves, since they were trained on all the code that exists.

I don't know what conclusion to draw from that. Maybe that there's no such thing as a free lunch, after all.

Re: AI made coding more enjoyable

#96
Yes AI has taken away the tedium, but a lot of that could already be overcome by leveraging your text editing tools well or with basic code generation (such as being able generate the skeleton of a class from an interface).

And there was something nice about still having to put in the manual work in those cases. It let me process what the code is actually doing and gave me the opportunity to internalize it in a way that just doesn't happen with AI. It also sort of gave me a thinking break where I was engaged at just the right level to let the thoughts about the more interesting parts float around in my head. With AI writing all the code, I feel like I'm either fully engaged with those thoughts or not engaged at all. And that's a bit of a problem because aha moments often happen when the idea is in that middle area of thought.

Re: AI made coding more enjoyable

#97
post #95
post #47

Earlier quoted context omitted.

LLMs have limited working memory, like humans, and most of the practices that increase human programming effectiveness increase LLM effectiveness too. In fact more so, because LLMs are goldfish that retain no mental model between runs, so the docs had better be good, abstractions tight, and coding practices consistent such that code makes sense locally and globally.

So are we basically saying that LLMs work most effectively on codebases that exhibit good quality coding practices, but are not themselves particularly good at creating such quality code themselves, since they were trained on all the code that exists. I don't know what conclusion to draw from that. Maybe that there's no such thing as a free lunch, after all.

The conclusion I draw is that LLMs need a human expert with some taste and agency to empower/supervise them. Just like human dev teams do.

Re: AI made coding more enjoyable

#98
I like taking my code, especially sql stuff and asking AI a better way to write this and it actually is making me better at SQL because of sql methods i didnt know i could do before. I also like it in vscode to just quickly do the redundant things with auto-complete. now i have 15 years of programming behind me. If i had AI starting out, it would be very bad, I would lack understanding and be a incompetent programmer.

Re: AI made coding more enjoyable

#99
"What I find annoying is repetitive stuff that's just typing"

..

"Where I can't trust AI is if it needs to copy paste / duplicate code"

???

AI takes away the "boring", "tedius" parts of coding for you, yet you at the same time don't trust it to even just duplicate code from one place to another?

Re: AI made coding more enjoyable

#100
post #26

I suppose, in exactly the same way instant / frozen food makes cooking more enjoyable. If it was just a chore that you had to do, and now it's faster, sure, grab that cup-o-noodles. Knock yourself out. Just don't expect to run a successful restaurant based on it.

The famously unsuccessful restaurant of McDonald's begs to differ
Post reply on HN