There are numerous improvements in Android O. A lot of them are in app improvements but here is a list of the ones focused on creating more fluid experiences.

Fluid experiences in Android O

It's pretty incredible what you can do on mobile devices today, and how easy it is to rely on them as computers in our pockets. In the O release we've focused on creating fluid experiences that make Android even more powerful and easy to use, and today we highlighted some of those:

  • Picture-in-picture: lets users manage two tasks simultaneously, whether it's video calling your friend while checking your calendar, or reading a new recipe while watching a video on a specific cooking technique. We've designed PIP to provide seamless multitasking on any size screen, and it's easy for apps to support it.
  • Notification dots extend the reach of notifications, a new way for developers to surface activity in their app, driving engagement. Built on our unique and highly regarded notification system, dots work with zero effort for most apps - we even extract the color of the dot from your icon.
  • Autofill with Google simplifies setting up a new device and synchronizing passwords by bringing Chrome's Autofill feature to Android. Once a user opts-in, Autofill will work out-of-the-box for most apps. Developers can optimize their apps for Autofill by providing hints about the type of data expected or add support in custom views.
  • A new homescreen for Android TV makes it easy for users to find, preview, and watch content provided by apps. Apps can publish one or more channels, and users can control the channels that appear on the homescreen. You'll be able to get started with creating channels using the new TvProvider support library APIs.
  • Smart Text Selection: In Android O, we're applying on-device machine learning to copy/paste, to let Android recognize entities like addresses, URLs, telephone numbers, and email addresses. This makes the copy/paste experience better by selecting the entire entity and surfacing the right apps to carry out an action based on the type of entity.
  • TensorFlow Lite: As Android continues to take advantage of machine learning to improve the user experience, we want our developer partners to be able to do the same. Today we shared an early look at TensorFlow Lite, an upcoming project based on TensorFlow, Google's open source machine learning library. TensorFlow Lite is specifically designed to be fast and lightweight for embedded use cases. Since many on-device scenarios require real-time performance, we're also working on a new Neural Network API that TensorFlow can take advantage of to accelerate computation. We plan to make both of these available to developers in a maintenance update to O later this year, so stay tuned!