37 lines
1.3 KiB
Markdown
Executable File
37 lines
1.3 KiB
Markdown
Executable File
# Roadmap (Alpha)
|
|
|
|
This roadmap is intentionally pragmatic: it prioritizes "time-to-first-success" and UX gaps before deeper architecture work.
|
|
|
|
## P0 - Config Persistence
|
|
|
|
Goal: `deskify update <id>` works without `--url`, and `deskify list --verbose` can show URL/backend.
|
|
|
|
- Store metadata under `~/.local/share/deskify/apps/<id>.json`
|
|
- Persist (v1):
|
|
- `id`, `name`, `url`
|
|
- `backend`, `browser_bin` (optional), `profile_scope`
|
|
- `fullscreen`, `no_decorations`, `user_agent` (optional), `width`, `height`, `dark_mode`
|
|
- `schema_version`
|
|
- Migration: older installs without metadata should be migrated on first update/build
|
|
|
|
## P1 - Distribution / Onboarding
|
|
|
|
Goal: install `deskify` without a full Rust/Tauri toolchain.
|
|
|
|
- Promote "install from GitHub Releases" in README
|
|
- Keep CI releases producing a Linux binary asset
|
|
- Enable community packaging (AUR `deskify-bin`, etc.)
|
|
|
|
## P2 - Shared Runtime Model (only after P0/P1)
|
|
|
|
Goal: reduce per-app build time and prerequisites for the `tauri` backend.
|
|
|
|
- Concept: shared runtime/launcher + per-app configs (no per-app compile)
|
|
|
|
## P3 - QoL
|
|
|
|
- `deskify list --verbose` improvements (URLs/backend/profile scope)
|
|
- Better FAQ: WebView vs Chromium backend tradeoffs
|
|
- Optional DE integrations (later): KDE/GNOME widgets/extensions
|
|
|