Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 31 additions & 0 deletions architectureDecisionRecords/0005MetaFramework.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
# Title

Meta Framework

## Date proposed

05-20-2026

## Context

React has been recommending for the past 3 years to use a meta framework. They have been pushing the community towards SSR and React Server Components. There are multiple meta frameworks that support these.

The top 3 are

1. [NextJS](https://nextjs.org/) (~36,000,000 weekly downloads) (55% positive sentiment)
1. [React Router v7 Framework Mode/Remix](https://reactrouter.com/) (~2,300,000 weekly downloads) (68% positive sentiment)
1. [Tanstack Start](https://tanstack.com/start/latest) (~38,000 weekly downloads) (No data for sentiment)

Tanstack start seems really promising, but it's still an RC. React Router v7 also looks really promising, but Remix has had years to build adoption and hasn't gained a lot of traction.

NextJS has been the industry standard for a long time, and it is used way more than the other frameworks, so it feels like the safest bet.

## Decision

When building new apps we will default to NextJS. If a frontend project doesn't make sense to build in NextJs, then we can build it with our old stack of Vite and React.

## Consequences

We won't be able to just push our NextJS apps up to an S3 bucket. We will have to deploy them. Engineers will have to learn the NextJS framework. We will have to worry about server and client side state.

Users of our products will get faster page load times due to SSR. We have the option of building our backend in typescript with NextJS.
Loading