Worse, Vercel is involved, and I literally don't remember anything good about that company.
I'd recommend to be very cautious with such news, and use older versions of React for the next couple of years.
https://react.dev/community/team
This announcement mentions they are separating business and technical governance, I suspect they are trying to limit Vercel's influence, and prevent them from taking it in a direction that only benefits them.
The maintainer of MSW has been screaming for years for people to drop jest [1]
[0] - https://github.com/faker-js/faker/blob/428ff3328b4c4b13ec29d...
The ultimate win, of course, would be to use the native Node test runner. See the sourse of the Node.js website - I think they have pulled it off despite running a Node.js app.
Until recently Jest had a bug that made it crash due to sl (yes, the famous steam locomotive) running under the hood. This gives a hint at, ahem, the sophistication of its architecture.
The project is long in its EOL, and the only reason for its use is inertia, the jQuery kind of it.
If you've ever used any other test runner, you'll find Node's is woefully inferior. I'd say "but maybe it will get better", except I've seen the maintainer responses to several issues, and it seems they are wedded to bad architectural decisions that keep it that way.
But I wasn't using a lot of Jest features anyway, generally preferred Mocha even during the height of Jest's popularity, and Node's test runner is sufficient for most of my needs (and Deno's starts to seem more and more the path forward as I come to prefer deno.json in a lot more types of projects than package.json).
I cannot for the life of me understand why anyone would intentionally pick it in 2025 unless there were serious constraints that forced them to.
And so in that sense, the answer is web components. I know everyone hates the API but it was intentionally designed to be a low level thing to build a developer experience on top of and the best implementation of that right now is Lit. It’s also a pleasure to use and incredibly lightweight and designed to become even more lightweight overtime as new capabilities come to the web like signals in JavaScript or talk of native templating etc.
There is another option that I really like for certain kinds of applications which is Flutter which might sound contradictory to my original point because it skips the idea of the DOM entirely and brings its own rendering engine to a canvas element.
But it’s set to be the first serious Wasm and WebGPU based UI library on the web and has no problems spitting out 120fps (this is before even they have actually added WebGPU support and their newest rendering engine to the mix by the way) while keeping its rendering engine to I think about 1.5-2mb in size. It also gets you a single codebase that will run literally anywhere as an AOT compiled app.
That’s to say nothing of Dart itself which I cannot even begin to describe what a huge improvement it is over JavaScript and Typescript. If you haven’t tried it yet, do yourself a favour.
I’m just making the point that JavaScript and the DOM are no longer the only players in town and when you’re not trying to mix abstractions of a document markup language and an application a whole lot of problems just disappear like does this look different in different browsers or can I use this new feature etc.. it just works… everywhere… on the web.. on a desktop.. on a phone or an IoT device.
But my point stands. I haven't seen a single job posting in 2 years that isn't using React, Vue, Angular, or (a tiny bit) Svelte. So Web Components just aren't being widely used, it seems.
The only exception is about job postings and react, I’m not making the argument in any way shape or form that that isn’t the overwhelming majority of jobs out there, but that wasn’t what I was talking about I was saying it’s an actively bad choice for a project in 2025.
I'm also not trying to be rude but you said a lot about React and offered a myriad of Google solutions to fix it, most of them got downgraded by Google itself and are on their way to the famous Google graveyard. Telling us about a problem and that Google knows how to fix it is indeed rude.
A Flutter and Firebase developer who goes by the name xeladu wrote that after a discussion with an unnamed “Google developer expert,” the developer wouldn’t recommend learning the language, but added it’s not time to port Flutter projects just yet.
“I’d rather say no to be totally honest. Only do it if you just want to play around. Then it will be fine,” xeladu wrote. “But becoming a serious professional Flutter developer could probably be a waste of time.”
https://techcrunch.com/2024/05/01/google-lays-off-staff-from...
I think maybe you might be mixing up some things here which is that there was once many years ago a version of Angular which was written in Dart which exists internally and powers AdWords but no longer exists publicly. Perhaps that is why you’re thinking of?