> The production build metric here would be ESBuild + a small amount of overhead from vite. (The HMR/dev build side of things is handled by Rollup instead with Vite)
I suppose we should wait until all tools move to Rust/Zig before learning modern front-end engineering to bypass the churn phase. Till then stick to plain HTML+CSS.
I've been using webpack for the last 7 years. Major version upgrades can be a pain, but broadly it's fine. People hopping frameworks every 6 months are causing their own problems.
I fully agree with the sentiment - but for me there's also that I could remove 26 (!) dev dependencies and minutes off each build, in a medium-sized app, when replacing Webpack with Vite - so it's worth being in the loop at least to some extent.
Yep, that makes total sense. We're also looking at a Vite migration in the near future, but also making sure that whatever we do, we're leaning more on web standards and common functionality, and avoiding as much bundler specific stuff as possible.
Back in the day is was cute that you could import GLSL fragments directly into your JS code. Now, not so much.
for the basics you can just use vite and nothing will change except what's underneath. it's only when you get fancy that you need to write your own plugins that would really be affected by this churn.
That’s completely backwards. Vite only uses rollup for production, which is why production build is slower. https://vitejs.dev/guide/why.html#why-not-bundle-with-esbuil... The plan is to replace both esbuild and rollup with Rust-based rolldown (fast, and supports rollup’s plugin architecture) eventually. See https://www.youtube.com/watch?v=hrdwQHoAp0M.