What’s new for developers in Android P

The initial developer beta of Google's forthcoming Android P operating system looks to entice developers with enhancements forKotlin coding, machine learning, and application compatibility.

The use of Kotlin as a supported language inAndroid Studio(via a plugin) lets developers improve the performance of their code, Google says.

For on-device machine learning, the Neural Networks API introduced in Android 8.1 will add nine capabilities: Pad, BatchToSpaceND, SpaceToBatchND, Transpose, Strided Slice, Mean, Div, Sub, and Squeeze. The C API runs computationally intensive machine learning operations.

To boost app compatibility with Android P, Google will gradually limit access over time to non-SDK-provided device interfaces. The Android P beta's updated SDK provides system images for testing on the Android Emulator and on various Pixel devices.

Other improvements in Android P include:

  • For security, a more consistent UI for fingerprint authentication.
  • More-accurate indoor positioning via Round-Trip-Time (RTT) APIs, with which apps can measure the distance to Wi-Fi access points. With knowledge of the distance to access points, device position can be calculated as granularly as one meter.
  • Enhanced performance and efficiency on all apps in the ART (Android Runtime).
  • New APIs in the Autofill Framework, to save time in filling out forms in password managers.
  • To assist with mobile payments, an implementation of GlobalPlatform Open Mobile API.
  • Display cutout support, to support full-screen content on new device screen sizes.
  • New APIs have been added to manage display of content onscreen, such as to support notches.
  • The  MessagingStyle feature for message notifications that highlights who is messaging and ways to reply.
  • A multicamera API, for access to streams simultaneously from two or more cameras. Developers can build features for devices that have dual-front or dual-back cameras such as seamless zoom and stereo vision. Also, developers can use this API to call a logical or fused camera stream and automatically switch among cameras.
  • Sessions parameters for cameras, to reduce delays after initial capture.
  • A surface-sharing capability, so camera clients can handle different use cases with no need to stop and start streaming.
  • APIs for display-based flash support as well as app-level image stabilization and special effects.

Where to download the Android P developer beta

You can download Android P system images or install the Android Emulator from the Android website .