I've been building loads of stuff with it for years and never experienced any of this. Sounds like a YOU problem.

multiple other comments sharing my experience and expanding on it.

If its so crappy how's it so popular?

The core React documentation explicitly points you to nextjs and says "this is how you should build react apps."

I mean, that's good enough for me.

They spend a lot of money on marketing.

I often read that Next.js sucks. Meanwhile I and many other devs I've spoken to IRL find it does what we need it to do without any issues. Ya'll just some haters.

Just because you didn’t see the issues other people are encountering, everybody else’s experiences are invalid?

I'm saying the impression you'd get of Next from this thread is that it literally does not work. It works fine.

For all my criticisms of it, I built two commercial apps in it. I worked around the issues and it was fine.

I've also built commercial apps in other stacks and they also have their warts.

What I've noticed from the other stacks, however, is that the frequency of entirely unnecessary issues is simply lower. React and NextJS aren't going anywhere and one can hope that these things will improve over time.

Ultimately, it's also a great employment guarantee, as companies will need people to maintain the apps that are constantly changing.

I think applying scepticism to Vercel and its motives is healthy, still.

See the various other comments for concrete examples of why nextjs sucks and the team at vercel is incompetent when it comes to auth, middleware, caching, and just generally maintaining a usable framework without brutal migrations and api breakages.

They have made egregious mistakes that go far beyond "move fast and break things" and well into "we should have the lawyers join this call".

Built 4 commercial projects with it now. Not had these issues.

I'd never trust it

Do you have to trust it? Have the AI built a list of all the invoices it can find in your email then a separate application reconciles that against your statements. The AI might miss something or invent something, but the mismatch list you get back should be short.

It wouldn't be perfect, but maybe it would be better than having to do it all manually.

I think many HN readers have a sensitivity to noise.

That doesn't fit his narrative so it's ignored.

Haha I'm not that dense. I'll give you counter examples. Bitcoin (supposed to revolutionize the world still waiting), Apple Vision Pro (supposed to change how we work, wear it on streets and airplanes, still waiting..)

Right, so why do you think vibe coding is digital cameras rather than bitcoin? Without that you haven't said anything, just made a prediction about the future without giving any reason to think it's correct.

With Bitcoin I immediately saw that it was a solution looking for a problem it still hasn't found.

NFTs and Web3 were a VC and hype powered rug-pull from the beginning.

LLMs on the other hand can solve problems. Not every problem, not perfectly. But you can solve actual things with them today.

Is there hype? Fuck yes there is. So much hype and snake oil. But there's a nugget of actual usefulness hidden in there.

Nobody's disputing that. The video creator himself has videos about AI coding techniques. It's "vibe coding" - using LLMs to generate code without reviewing or understanding it - that is at issue.

It all depends on the goal IMO.

If you're building "production ready" stuff with logins for Other People and god forbid taking payments, yes you definitely must understand what the code does.

But when I'm building a tool that tags random meme reaction .webm files and finds duplicates from them, I couldn't care less what the quality is or if I understand what it does.

I can easily run it, observe if it does what I want and Vibe harder if it doesn't.

tl;dr vibing prod code = bad. vibing personal tools = good.

I wear my AVP on airplanes. Only thing holding it back is the price…

Yes of course! Who cares if it looks worse than a human can do, it was free and instant . And that's all that matters in our AI future.

I'm tired of all the Next.js slander. It works great.

So sick of all this now. Honestly any time I go online to learn about AI writing code, I end up knowing less. One person will say 'Lets not fire all our devs, AI is another tool not a replacement'. Then you get another guy saying 'Lol yeah right just wait till SuperAI v3.6 comes out, then we'll all be fired'

What the hell is wrong with you people? These are people's livelihoods.

Well I guess I'm in the former category but what do you expect us "the people" to do about it? I don't think you should halt progress just because people would lose jobs because people create and do jobs because there is something that needs to be done. If we keep them around just so people have jobs then we're running an adult kindergarten which makes no sense to me. Besides, there will be jobs that can't be done by AI for a long time to come and the current AIs are relatively crappy compared any average software dev. I wonder why this is such a popular topic even, to me it feels like if you'd use one of these LLMs for more than a week you'd start to see why it's incapable of replacing anything major.

> What the hell is wrong with you people? These are people's livelihoods

Do you think about lamp lighters the same way?

If you're able to compare software devs with lamp lighters, you've truly bought in to the AI hype.

If lamp lighters also build and maintain the lamps, produce the town gas, build and maintain the gas pipes, clean the streets, you sure as hell shouldn't fire them, because they invented self lighting lamps.

Yes? And train wagon brake operators.

Social upheaval sucks for those involved.

I dont think he's getting enough hate for using AI to do traditionally creative work. Thanks for making the future worse asshole.

You can safely ignore any blog post with 'The death of...' in it's title.

As you say, companies telling us their products with dominate the future is nothing new. The next big job killer is always 5 years away, apparently.

> The next big job killer is always 5 years away, apparently

It really is, and really always has been. At least, when it comes to software.

What's always been hard, is knowing which of the many predictions about the real job killer will be correct, so that you can train for it.

In 2009, I was expecting that by 2019 the world would have widespread availability of self driving cars without steering wheels, and not just some geofenced examples and a promise of more "real soon now honest" as we're still at in 2024.

In 2019, I wasn't expecting LLMs to be even remotely as good as they turned out to be by 2022, let alone 2024.