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

Don't fall into the anti-AI hype

antirez.com

Post reply on HN
Print view View original

501–510 of 1001 posts

Re: Don't fall into the anti-AI hype

#501

Post by visarga »

phicoh wrote: post #198
embedding-shape wrote: post #95

> But what was the fire inside you, when you coded till night to see your project working? It was building. I feel like this is not the same for everyone. For some people, the "fire" is literally about "I control a computer", for others "I'm solving a problem for others", and yet for others "I made something that made others smile/cry/feel emotions" and so on. I think there is a section of programmer who actually do…

The problem I see is not so much in how you generate the code. It is about how to maintain the code. If you check in the AI generated code unchanged then do you start changing that code by hand later? Do you trust that in the future AI can fix bugs in your code. Or do you clean up the AI generated code first?

You don't just code with AI, you provide 2 things

1. a detailed spec, result of your discussions with the agent about work, when it gets it you ask the agent to formalize it into docs

2. an extensive suite of tests to cover every angle; the tests are generated, but your have to ensure their quality, coverage and depth

I think, to make a metaphor, that specs are like the skeleton of the agent, tests are like the skin, while the agent itself is the muscle and cerebellum, and you are the PFC. Skeleton provides structure and decides how the joints fit, tests provide pain and feedback. The muscle is made more efficient between the two.

In short the new coding loop looks like: "spec -> code -> test, rinse and repeat"

Re: Don't fall into the anti-AI hype

#502

Post by simonw »

jfengel wrote: post #480
zeroonetwothree wrote: post #311

Earlier quoted context omitted.

You’re right of course. For me there’s no flow state possible with LLM “coding”. That makes it feel miserable instead of joyous. Sitting around waiting while it spits out tokens that I then have to carefully look over and tweak feels like very hard work. Compared to entering flow and churning out those tokens myself, which feels effortless once I get going. Probably other people feel differently.

I could imagine a world where LLM coding was fun. It would sound like "imagine a game, like Galaxians but using tractor trailers, and as a first person shooter." And it pumps out a draft and you say, "No, let's try it again with an army of bagpipers." In other words, getting to be the "ideas guy", but without sounding like a dipstick who can't do anything. I don't think we're anywhere near that point yet. Instead we'…

Prompt one:

  imagine a game, like Galaxians but using tractor trailers,
  and as a first person shooter. Three.js in index.html
Result: https://gisthost.github.io/?771686585ef1c7299451d673543fbd5d

Prompt two:

  No, let's try it again with an army of bagpipers.
Result: https://gisthost.github.io/?60e18b32de6474fe192171bdef3e1d91

I'll be honest, the bagpiper 3D models were way better than I expected! That game's a bit too hard though, you have to run sideways pretty quickly to avoid being destroyed by incoming fire.

Here's the full transcript: https://gisthost.github.io/?73536b35206a1927f1df95b44f315d4c

Re: Don't fall into the anti-AI hype

#503

Post by epolanski »

burgerone wrote: post #416

There's this infinite war between the two opposing sides. "It's going to change programming forever" vs "Why not just use your brain". I much prefer option two for all the good reasons. Saying that AI is awesome doesn't actually adress all its issues.

There's no war, it's just that the internet pushes the more controversial narratives up.

The overwhelming majority of users use and see the benefits of AI and at the same time are fully aware that you won't move software by copy pasting a jira task and lots of thinking is involved into planning and reviewing the changes.

Re: Don't fall into the anti-AI hype

#504

Post by epolanski »

edg5000 wrote: post #98

> state of the art LLMs are able to complete large subtasks or medium size projects alone, almost unassisted, given a good set of hints about what the end result should be No. I agree with the author, but it's hyperbolic of him to phrase it like this. If you have solid domain knowledge, you'll steer the model with detailed specs. It will carry those out competently and multiply your productivity. However, the quality…

Didn't somebody built a rather decent and fully compliant html parser by copy pasting 8000+ tests from another project?

That qualifies as a good set of hints about what the end result should be.

Re: Don't fall into the anti-AI hype

#505

Post by lifetimerubyist »

UBI will never happen because the people in power done want it.

Who is going to control AI? The people in power obviously. The will buy all of the computers so running models locally will no longer be feasible. In case it hasn’t been obvious that this is already happening. It will only get worse.

They will not let themselves be taxed.

But who will buy the things the people in power produce if nobody has a job?

This is how civilization collapses.

Re: Don't fall into the anti-AI hype

#506

Post by epolanski »

adityaathalye wrote: post #27

Don't fall into the "Look ma, no hands" hype. Antirez + LLM + CFO = Billion Dollar Redis company, quite plausibly. /However/ ... As for the delta provided by an LLM to Antirez, outside of Redis (and outside of any problem space he is already intimately familiar with), an Apples to Apples comparison would be he trying this on an equally complex codebase he has no idea about. I'll bet... what Antirez can do with Redis…

I'm not sure the blog post goes in the opposite direction of what you say, in fact he points out that the quality of the output depends on the quality of the hints, which implies that quality hints require quality understanding from the user.

Re: Don't fall into the anti-AI hype

#507

Post by Krssst »

burgerone wrote: post #416

There's this infinite war between the two opposing sides. "It's going to change programming forever" vs "Why not just use your brain". I much prefer option two for all the good reasons. Saying that AI is awesome doesn't actually adress all its issues.

Plus, a big part of an engineer's job is understanding what the system is doing and how. And correctness. Letting LLMs write everything without a deep review that would take at least half the time it would take to design and write the thing manually does not seem to meet that goal.

Re: Don't fall into the anti-AI hype

#508

Post by kgdiem »

zeroonetwothree wrote: post #311
embedding-shape wrote: post #95

> But what was the fire inside you, when you coded till night to see your project working? It was building. I feel like this is not the same for everyone. For some people, the "fire" is literally about "I control a computer", for others "I'm solving a problem for others", and yet for others "I made something that made others smile/cry/feel emotions" and so on. I think there is a section of programmer who actually do…

You’re right of course. For me there’s no flow state possible with LLM “coding”. That makes it feel miserable instead of joyous. Sitting around waiting while it spits out tokens that I then have to carefully look over and tweak feels like very hard work. Compared to entering flow and churning out those tokens myself, which feels effortless once I get going. Probably other people feel differently.

I feel the same way often but I find it to be very similar to coding. Whether coding or prompting when I’m doing rote, boring work I find it tedious. When I am solving a hard problem or designing something interesting I am engaged.

My app is fairly mature with well established patterns, etc. When I’m adding “just CRUD” as part of a feature it’s very tedious to prompt agents, reviewing code, rinse & repeat. Were I actually writing the code by hand I would probably be less productive and just as bored/unsatisfied.

I spent a decent amount of time today designing a very robust bulk upload API (compliance fintech, lots of considerations to be had) for customers who can’t do a batch job. When it was finished I was very pleased with the result and had performance tests and everything.

Re: Don't fall into the anti-AI hype

#509

Post by christophilus »

Diti wrote: post #462
jstummbillig wrote: post #453

Earlier quoted context omitted.

There is no hard part. The anti-AI position has simply become trite. The idea is that agentic coding does not work. Today, it does work.

It only works for languages and frameworks that are already in the training data (duh). It still is mostly useless when you need to create something from scratch in an unstable language. That, and you can’t also get the amazing results if you’re poor or have bad internet.

Not true. I built some tools in Hare, which almost certainly isn’t in the training data to any significant extent. It was more work than having it build Go or Rust, but it got it done. It had to curl the docs a fair bit.

Re: Don't fall into the anti-AI hype

#510

Post by simonw »

csmpltn wrote: post #414
simonw wrote: post #160

Earlier quoted context omitted.

This is a pretty common position: "I don't worry about getting left behind - it will only take a few weeks to catch up again". I don't think that's true. I'm really good at getting great results out of coding agents and LLMs. I've also been using LLMs for code on an almost daily basis since ChatGPT's release on November 30th 2022. That's more than three years ago now. Meanwhile I see a constant flow of complaints fro…

So where’s all of this cutting edge amazing and flawless stuff you’ve built in a weekend that everybody else couldn’t because they were too dumb or slow or clueless?

I wouldn't call these flawless but here you go:

- https://github.com/simonw/denobox is a new Python library that gives you the ability to run arbitrary JavaScript and WASM in a sandbox provided by Deno, because it turns out a Python library can depend on deno these days. I built that on my phone in bed yesterday morning.

- https://github.com/simonw/pwasm is a WebAssembly runtime written in pure Python with no dependencies, built by feeding Claude Code the official WASM specification along with its conformance test suite and having it hack away at that (again via my phone) to get as many of the tests to pass as possible. It's pretty slow and not really useful yet but it's certainly interesting.

- https://github.com/datasette/datasette-transactions is a Datasette plugin which provides a JSON API for starting a SQLite transaction, running multiple queries within it and then executing or rolling back that transaction. I built that one on my phone on a BART (SF Bay Area metro) trip.

- https://github.com/simonw/micro-javascript is a pure Python, no dependency JavaScript interpreter which started as a port of MicroQuickJS. Here's a demo of that one running in a browser https://simonw.github.io/micro-javascript/playground.html - that's my JavaScript interpreter running inside Python running in Pyodide in WebAssembly in your browser of choice, which I find inherently amusing.

All of those are from the past three weeks. Most of them were built on my phone while I was doing other things.

Post reply on HN