Hacker News Reader

Hacker News

Comment by aloukissas | Hacker News Reader

Full thread

aloukissas

Prediction: Google buys Vercel, kills Angular, steps further into e-commerce

View on HN

… and people would look into https://svelte.dev which might be everything you need.

They don't have enough funding. Next.js and Gatsby are in an entirely different weight class compared to Sapper.

https://sapper.svelte.dev/

And... Sapper is dead btw.

Rich said it recently in a video. The Svelte team is working on the next big thing combining Svelte with Snowpack. The project is called SvelteKit.

https://www.youtube.com/watch?v=qSfdtmcZ4d0

I think retiring Sapper was the right decision. I have a feeling Rollup will meet a similar end too.

They both seemed like fine pieces of software. They just don't seem offer separate much their counterparts nowadays that warrants continued feature development.

Rollup isn't going anywhere! When you poke behind the scenes of newer build tools, like Snowpack and Vite and the custom compiler that powers Remix Run, you'll find Rollup doing a lot of the heavy lifting. And it's the preferred choice of most libraries. There's a reason for that :)

Thanks for the correction, Rich.

I just watched your presentation at Svelte Summit and I'm looking forward to how Svelte progresses.

Maybe this can save someone some confusion: When I searched the first result for SvelteKit seems to be another project that is dead.

Should be called SveltePack IMO :)

The project hasn't really been publicly released yet. The repository is still private. The package is up on npm, but isn't ready for widespread adoption yet. I'd recommend sticking with Sapper for now.

https://www.npmjs.com/package/@sveltejs/kit

"Sapper is dead" is not really accurate. SvelteKit is Sapper + Snowpack and there will be a migration path.

SvelteKit will replace Sapper which will never reach 1.0. Development on Sapper will stop (or probably has already stopped).

Or even better, Vue.js

I am not sure I buy into all the svelte hype, and even if claims about it's speed/build size are true - the difference isn't enough to outweigh the enormous advantage vue has in terms of a more mature ecosystem/tooling.

Could be. Angular growth has stalled for the past year or so. Polymer is pretty much dead.

OTOH hand why would Google invest in a React-based solution? Google could start from scratch an create the next best thing if they wanted to. To me it's pretty clear the future is a compiler based approach like Svelte, with support for SSR, partial hydration, static generation, and dynamically loading smaller parts of a page (eg: AJAX + hydration). All with a kickass DX.

They could buy Vercel & Next with the intention of rolling their own React replacement to swap in later.

My bet is because of the ecommerce part of the platform (integration /w headless ecommerce companies like BigCommerce).

If Google wanted to takeover React, the solution is at the browser level. Fix rendering and state management with native language built ins, and voila, we are out of the stupid frontend hell.

Wtf is hydration? Is that like when websites used to render json payloads on the page on first render?

This field is getting scammy with its nonsense. Create one more stupid word, and I swear I will lose my god damn mind.

If that happens I'm out, especially if they bundle it other Google products

Roll it into Firebase maybe

Reply on news.ycombinator.com