[Existenceblinks]

> Crucially, none of the React alternatives solve this

FFS. The context is the web, nobody solves graphic problem like gaming for you, or need to. At this point it doesn't matter because HTML/DOM stuff is not going to suffice what you are talking about either.

I honestly think after reading 50% of the article is all about bragging knowing history. I'm not old but I was there too. WIN32, MFC, QT, 8086 assembly whatever. I got a god damn computer engineering degree too but doesn't make me smarter, having better vision, knowing solution better than random ppl on the internet.

I'm not sure what's the point of React here nor "Saving React". Why it needs to be saved in which sense.

Do you mean, saving the web from React?

[friendlyHornet]

Yeah, I stopped halfway through the article.

Don’t talk down and condescend to your audience for no reason

I think the article should be rewritten and all the ranting removed

It’s also difficult to follow where the author‘s going with all the side rants and unrelated remarks

And yes, that is considered ranting, contrary to what the author claims in the beginning of the article

[technion]

I do agree with you, but to play devil's advocate: This article is currently #9 trending on the front page. I've meanwhile seen similar critiques of React written in a different style, and seen them not get a vote or comment. Perhaps those rants do serve a purpose?

[nindalf]

Probably getting upvoted based on the catchy title alone.

[matheusmoreira]

Can't wait for the "ok boomer" article refuting this rant.

[nwienert]

I didn't quite make the cut as a Zoomer, but I think what I wrote today is:

https://tamagui.dev/docs/intro/compiler

[inductive_magic]

>I'm not sure what's the point of React here nor "Saving React". Why it needs to be saved in which sense.

I read it as “svelte et al gain traction due to inferior younglings, therefore react is a sinking ship, let’s not let it sink because the tooling I use is the adequate choice (despite having to be saved)”

[Existenceblinks]

Yeah. The first paragraph is that point, but the follow up paragraphs are more of general sense of building UI.

The other frameworks or libs lately move towards non virtual dom and reactivity. Solid, Preact, Svelte, Vue, Lit are superior engineering wise. But React is about the whole economy (unbelievable VCs put 100+ millions on that), so more libs more stuff created, rich ecosystem, poor engineering.

[wokwokwok]

> FFS. The context is the web, nobody solves graphic problem like gaming for you, or need to…

I don’t think you understood the article.

The point he was making is that a good user interface is hard to build and is “app like”, with all the weird edge cases.

…and that, for better or worse:

1) react is the best at doing that.

2) react is (frustratingly) not making itself better at it, and is instead prioritising weird stuff like server side rendering, that is pretty niche in terms of value.

3) react competition is basically just “still the same shit web ui” but easier to build and with nicer tooling.

So, I mean, yeah it’s a rant.

…but hey, I think you’re failing if you walk away from it going “wtf was that?”.

He’s right.

Making “app like” websites is pretty hard; it’s very hard with some frameworks.

Apps are better, in almost every way.

That sucks.

No one except react was really even fighting that battle, and they seem to have lost interest.

That kind of sucks too.

Does react suck? Maybe.

…but, tldr; if you’re gonna take a big dump on react from a grey height, at least do so from a position of thoughtfulness.

“No, I’m happy with building a web ui, I don’t need slick user interactions or an interface design rather than just a form”

Cool. That’s fine. The MVP will be quick to build.

Not everyone needs to be figma.

…but pick the tool for the job; and some of the new frameworks are very finely crafted tools for a very specific job; and that job is building a generic web ui.

…but hey, the web is the context right?

So that’s fine, if you don’t care.

[Existenceblinks]

It's either DOM or Canvas as foundation we have right now in browser. Not seeing economy around WebGL that much. He talks about GPU multiple times. And these aren't thing related to React at all. React is not "the best at doing that", so I disagree here .. thoughtfulness is that it's repetitive thing to say on HN, people disagree with virtualDOM for a long while, Signal (reactivity) isn't new idea either. More thoughtful thing to say is get "DOM part" and "Template Instantiation" to work so that we have an efficient update mech basis to rely on.

I wanted to build browser app without DOM too, but this is what we have! People will complain about accessibility and all default behaviour goodness that will be thrown away.

I care, we have to start with the right basis which is not React's.

[LtWorf]

But why can't we make an app if we need an app?

[mort96]

Sometimes, you want to be able to support more than one OS.

[LtWorf]

You mean like download qt and make an app that compiles on everything and integrates much better than a webpage while also being much faster and smaller?

[efdee]

Don't expect me to download your .exe if you're solving a problem that can be solved using a web app.

[LtWorf]

Don't expect me to care if my exe solves your problem but you are too stubborn to use it.

[tuukkah]

He does not mention gaming once though. The web quickly got to the point where UIs are possible, but lately there's seemingly little or no progress in making app-quality UI possible let alone easy.

In this sense, we might be in a situation where the floodgates are about to open, but who and what is able to make it happen?

[jeswin]

> I honestly think after reading 50% of the article is all about bragging knowing history.

This is why I can't read most tech articles these days. Won't get to the point without a long detour into the background of the problem; as someone in my forties, it's hard to justify the expense.

[Existenceblinks]

I kinda get him having to drag these historical UI to draw some points but ended up having no critical points for current context and title. I guess it's info for "Get in Zoomer" part but failed to demonstrate "We're Saving React". If a zoomer read that, what should they do, what's the point to get start here, and should they save React or start something new?

[cyber_kinetist]

(As a zoomer,) I know that React is designed as the way it is because it's limited by the DOM model for all rendering and user interaction, whereas for native apps you have much more freedom in terms of design space of GUI toolkits (from a completely retained model like Qt to highly immediate ones like IMGUI).

The disappointment of the article is that he talks about how the Mac OS X apps were the pinnacle of desktop applications, but he never talks about how people have achieved this. Did the Cocoa API made some special design decisions (as opposed to MFC or Qt) that enabled devs to make more reactive apps?

[Existenceblinks]

You know it well (DOM, Retained | Immediate). There's nothing to do with saving react here. Talking about browser app specially, I'd keep eye on standard like Template Instantiation and the coming up Symbol as Map's key. It's closed to the recent "Signal (reactivity)" trend. A way to update part of tree more efficiently without re-render-the-whole-thing. As long as we still use DOM, it's not going to be more revolutionary than this.

[atom_arranger]

The only possible Revolution I see is not using the DOM. JS + WASM is fast enough, the DOM is not fast enough, or good enough, enough here meaning enough for creating highly dynamic 60FPS experiences. Main issue with leaving it behind is A11Y though.

[lawgimenez]

I only made it to the first paragraph.