Skip to content

desktop-roadmap.md

Latest commit

 

History

History
76 lines (50 loc) · 2.59 KB

desktop-roadmap.md

File metadata and controls

76 lines (50 loc) · 2.59 KB
Raw

Desktop Roadmap

This note summarizes the current desktop design, why it is maintainable, and what should come next.

Current Design

Schema Studio desktop currently uses:

  • a Tauri shell for the native window and installer packaging
  • a packaged Python backend sidecar for Light Mode
  • the existing local HTTP backend/frontend split

That design has worked well so far because it keeps the desktop layer thin:

  • Tauri owns startup and shutdown
  • the backend stays responsible for application behavior
  • packaged builds remain self-contained for end users

Maintenance Assessment

The current design is good enough for ongoing maintenance because:

  • the desktop launcher is now organized around a small mode contract
  • the Windows and Linux packaging paths share the same sidecar model
  • the packaged backend is isolated from user Python installations
  • shutdown behavior is explicit instead of relying on implicit shell cleanup

The main maintenance risk is documentation drift between:

  • repo README
  • desktop usage/build notes
  • future mode-extension notes

When desktop behavior changes, those docs should be updated together.

Near-Term Priorities

  1. Keep Light Mode stable across Windows and Linux packaging flows.
  2. Add CI coverage for desktop bundle production where practical.
  3. Keep the packaged runtime self-contained and avoid introducing manual setup for users.
  4. Decide how packaged builds should handle schema packages that are normally first fetched from the internet.

Schema Package Strategy

The next important product decision is how to handle packages such as nomad-simulations that may need an online fetch or refresh step.

Options to evaluate:

  1. keep shipping a bundled snapshot only
  2. allow an explicit in-app online bootstrap/update flow when internet is available
  3. support a hybrid model with a bundled fallback plus an opt-in online refresh

Whatever path is chosen should preserve:

  • a working offline-first install
  • a clear user experience when the internet is unavailable
  • deterministic behavior for packaged desktop releases

Medium-Term Priorities

  1. Align desktop versioning with the main package version if release cadence becomes shared.
  2. Add release automation for Windows and Linux desktop artifacts.
  3. Improve packaged update strategy for the bundled schema snapshot.

Future Modes

If heavier modes are introduced later, prefer:

  • one packaged backend entrypoint per mode
  • one health endpoint per mode
  • one launcher descriptor entry per mode

Avoid making the Tauri shell directly orchestrate many unrelated helper services unless there is a strong reason to do so.