We've opened a Request for Comments (RFC) to finalize a major shift in the Framework M frontend and mobile strategy. Since Framework M is in pre-alpha with no legacy users to migrate, we have a unique window to bake "Total Convergence" into the foundation of the project now.
The Proposal: One Codebase, Any Screen
The core idea is to eliminate the "split-brain" model (where Web and Mobile are separate) and instead adopt a unified architecture:
- Tamagui Everywhere: Moving away from Ant Design to use Tamagui for both the Web Desk and Mobile apps for a "zero-cliff" design system and high-performance styling.
- Headless Logic Sharing: Extracting
@framework-m/desk/clientto house all Auth, CRUD, and Real-time logic in a platform-agnostic way. - Unified Tooling: Leveraging
onestack.devto use Vite as a universal bundler for both Web and Native, replacing Metro. - Local-First: Establishing an offline-sync strategy using SQLite and NATS CloudEvents from day one.
🛠 Why now?
Because we are pre-alpha, we can pivot to a high-performance, universal stack without the pain of migration later. We want to ensure the architecture is "Universal by Default" before we hit Alpha.
Join the Discussion
We are looking for feedback on the technical stack, the choice of Tamagui, and the headless core approach.
Read the full RFC and drop your thoughts here:
👉 Merge Request #146: Mobile Strategy - React Native & Tamagui Convergence
What are your thoughts on "Tamagui All The Way"? Is Vite-for-Mobile the right move for our DX? Let's discuss.