Frontend Tooling / Framework #214
bestickley
started this conversation in
Polls
Replies: 1 comment
-
Added $0.02 to the discussion in the issue (#221): |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Currently, Green Boost templates created from
gboost
CLI use Vite for frontend build tooling. Vite offers an excellent DX, it's fast, pluggable, and widely supported by the community. It's used by many JS frameworks for tooling and even Angular is starting to build their tooling on top of Vite so many devs have a vested interested in Vite's success which is good. However, Vite + React is best for building SPAs, not SSG (content sites) or SSR (e-commerce) sites. There isn't widely agreed upon solution for SSG vs SSR. I want GB's frontend tooling / framework to be flexible to support those types of web sites/apps. The de-facto solution for those types of apps built with React is Next.js. Next.js coins themselves as the "The React Framework for the Web". They have a large community and Next.js 13 is starting to take advantage of the new React Server Components. Next.js also provides a built in routing solution which is compelling. Currently GB uses React Router which has worked well, but your framework can optimize your website more if it's integrated into the router. Next.js 13.4 also just stabilized the "app directory feature" which adds many new features.Next.js tooling doesn't currently provide as great of a DX as Vite's but this will likely change as turbopack stabilizes.
Next.js SSR sites are built to be deployed on Vercel's platform and isn't super simple to openly deploy to AWS. OpenNext.js is changing this with an easy way to deploy with the CDK to AWS with CloudFront.
Best Reasons to use Next.js:
Best Reasons to stick with Vite:
React.lazy
with React Router data routers) to improve performance/web vitalsRelevant Links:
3 votes ·
Beta Was this translation helpful? Give feedback.
All reactions