OpenHarmony for Next Gen Mobile
summary
OpenHarmony 5.0 drops Android compatibility layer, so apps must be rewritten for the new Ark/RTS (a stricter TypeScript) and signed per-distribution; Huawei funds a top 5,000 apps port (about 4,000 underway). Porting Servo (≈240k Rust lines, 700+ deps, millions of vendored lines) proves feasibility: build requires target aarch64_unknown_linux_ohos, explicit linker env for cargo, cmake/corrosion workarounds, bingen/clang path fixes, font/EGL and RQX window glue; result: Servo renders on device but lacks fling and full WebGL, and needs further bindings.
AI-written summary · about
chapters (13)
- 0:00Introduction to OpenHarmony
- 1:08OS evolution and strategy
- 3:24Distributions and ecosystems
- 6:12App porting and HarmonyOS Next
- 10:05Development environment
- 13:21Security and signing
- 15:07App architecture and layout
- 17:54The RTS language framework
- 20:49Native code and integration
- 24:20Compiling Rust for OpenHarmony
- 28:59Porting the Servo engine
- 36:52Summary and future prospects
- 43:35Q&A session
related talks
20mChallenges in Porting Android to RISC-V
39mAndroid Open-source repositories 101
18mProgress in Porting Android onto RISC-V: Testing, Performance and Open Source
35mThe Ugly Truths about HTML5 for Mobile Apps
30mShipping production ready Rust for Android
31mAndroid's CHRE and Open Source Frameworks
44mC++, iOS, Android — Writing Cross‑Platform Applications
54mPanel Discussion: Why the Time Has Come To Disrupt Mobile Application Development