Research-Stack/5-Applications/linear-native-tauri
2026-05-08 15:04:30 -05:00
..
src Remove reqwest to fix ring linking errors; simplify dependencies 2026-05-04 19:27:44 -05:00
build-and-install.sh Add ENE research topic candidates, compression-core, notion-native-tauri, linear-native-tauri 2026-05-04 19:08:20 -05:00
build.rs Add ENE research topic candidates, compression-core, notion-native-tauri, linear-native-tauri 2026-05-04 19:08:20 -05:00
Cargo.lock Fix Dependabot dependency alerts 2026-05-08 15:04:30 -05:00
Cargo.toml Remove reqwest to fix ring linking errors; simplify dependencies 2026-05-04 19:27:44 -05:00
README.md Add ENE research topic candidates, compression-core, notion-native-tauri, linear-native-tauri 2026-05-04 19:08:20 -05:00
tauri.conf.json Fix Tauri builds: navigate API, unused imports, build config 2026-05-04 19:22:01 -05:00

Linear Native (Tauri)

Lightweight native Linux wrapper for Linear Web, replacing the laggy Electron build.

Why

The official linear-desktop-git (Tauri wrapper around linear.app) has ~1.5s input lag on NVIDIA/Wayland setups. This build strips the bundled frontend and loads linear.app directly in a native WebKitGTK WebView with GPU compositing.

Features

  • Multi-threaded: Tokio runtime with 4 worker threads + background prefetch
  • GPU-accelerated: Uses system WebKitGTK WebView (Mesa/Vulkan compositing)
  • No bundled frontend: Loads linear.app directly — always current
  • Native Rust: Compiled binary, not interpreted JS
  • Low latency: Input events go straight to WebKit, no intermediary Tauri frontend layer

Build

cd "/home/allaun/Documents/Research Stack/5-Applications/linear-native-tauri"
cargo build --release

First compile: ~10-20 minutes.

Install via aurutils

aur sync --no-view linear-native-tauri
sudo pacman -S --noconfirm linear-native-tauri

Launch

linear-native-tauri

vs linear-desktop-git

linear-desktop-git Linear Native (Tauri)
Frontend Bundled Tauri + custom UI Direct linear.app in WebView
Input lag ~1.5s (event routing overhead) Native WebKit event loop
Binary size ~30 MB ~15 MB
RAM idle ~200 MB ~80 MB
GPU Software fallback on NVIDIA Native Mesa compositor
Threads Single main thread Tokio multi-threaded pool