Google on Wednesday launched the first developer preview of Android P, the next major iteration of its operating system meant to succeed Android 8.1 Oreo. While the final, stable build of the software is likely to include a wide variety of features that still haven’t been announced, the initial experimental version of Android P already reveals many new functionalities users can expect from Google’s new take on a contemporary mobile experience, provided app developers opt to utilize them.
Possibly the most controversial addition is the new lineup of “display cutout APIs” meant to help developers create Android implementations for devices with screen notches akin to those seen on the Apple iPhone X and the recently announced ASUS ZenFone 5. The tools will allow software creators to easily identify non-functional areas of targeted device displays and program their user interfaces around them. Programming such interfaces doesn’t require a testing unit with a physical notch as the first developer preview of Android P is capable of simulating such a design cue. To do so, enable the system’s “Developer options” and locate the “Drawing” section on the following menu where you’ll find a corresponding functionality that will allow you to create a digital notch and adjust its size by pixels.
Android P is also coming with a broad range of notification improvements, as well as new functionalities meant to enhance the operating system’s notification management, particularly quick replies introduced with API level 24 that debuted with Android 7.0 Nougat. All messaging notifications can now include images, which is a feature that some apps such as WhatsApp have already been supporting independently for some time now, though native integration should make it much more widely available. Android P also allows developers to easily integrate saving notification replies as drafts and provides them with a new class that can be leveraged for quickly identifying notification participants with avatars and uniform resource identifiers. Messaging notifications in Android P can also natively determine whether particular chats are of a one-on-one or group variety and label them accordingly, whereas messaging notification actions such as replies, deletes, and full-screen toggles can now be given semantic meaning. The latter point has yet to be elaborated upon but likely has implications for machine learning and general artificial intelligence applications. Finally, Google’s suggested replies service is also being integrated into Android’s native Smart Reply functionality and will hence be available to third-party developers.
The initial developer preview of Android P introduces indoor positioning support utilizing the Wi-Fi Round-Trip-Time (RTT) protocol which leverages specific hardware to track a compatible device in a closed area based on its distance to various Wi-Fi access points. Apps with a permission to track host’s location can take advantage of the functionality, with the solution itself working even if the device isn’t connected to any Wi-Fi access point. The method still isn’t two-way and doesn’t allow devices like routers to read the determined position of a compatible smartphone or tablet so as to ensure a certain level of privacy, the Alphabet-owned tech giant said. Wi-Fi RTT tracking requires at least three separate APs and is accurate within one to two meters (3.23 feet) on average, according to the Android P team. While Google is expected to demonstrate the capabilities of the technology in the near future, the company is already predicting a wide variety of possible applications ranging from simple location service improvements to highly accurate voice controls, such as the ability to tell your phone to “unlock this door” when you’re standing near it. Augmented reality is another field that could benefit from Wi-Fi RTT tracking, whereas the hardware requirements of the solution indicate it will likely find the most usage in environments equipped with mesh networks.
Multi-camera support is now part of the package, allowing apps to access viewfinders or streams of two or more physical sensors. Google claims the ability to separate streams at an OS level should allow developers to deliver better bokeh-generating software, more seamless zoom functionalities, and various other features meant to enhance the mobile photography experience they’re striving to create. Support for stitched streams is also included, both in the sense that Android P can display a feed that switches between different sources based on some logical operators or literally show a fused image, with that functionality being yet another technology that may encourage developers to experiment with mobile cameras more aggressively going forward. Android P finally offers native support for display flashes that can be used for illuminating selfies and can provide third-party apps with optical image stabilization timestamps to accelerate their own optimization efforts. System-level compatibility with USB and UVC cameras is also included but will be hardware-dependent, i.e. not available on all smartphones and tablets.
Android P is being positioned to improve on the Notification Channels introduced by Android 8.0 Oreo with the goal of allowing users to granularly control the priority of their notifications based not just on their app of origin but also their type. Google is now primarily seeking to streamline existing functionalities instead of adding entirely new ones, with one such change coming in the form of an ability to block channel groups. E.g. while Android 8.0 Oreo allowed you to block all promotional notifications, you had to tell it to do so for each app individually, whereas that problem may now be eliminated unless many developers of your regularly used apps insist on custom channels or completely ignore the functionality. Apps will now also be able to determine the current channel group settings that affect them so that they don’t waste resources on attempting to send notifications guaranteed to get blocked. This two-way communication is being further deepened with the introduction of broadcast intent types meant to notify apps of blocked channels and provide them with data to alter their behavior.
Google introduced new tools for decoding images and drawing WebP and GIF animations that developers can leverage in order to make user interfaces of their apps more seamless. Android P debuts support for the High Efficiency Image File Format (HEIF) meant to optimize still images more aggressively but without major trade-offs on select devices. The company is encouraging developers to give the specification a go as it could significantly reduce the size of image-heavy apps, regardless of whether they’re pre-loaded with imagery or designed to download it on the go. The system’s media backend has also been ennobled with the addition of the High Dynamic Range (HDR) VP9 Profile 2 which will enable HDR streams on YouTube, Google Play Movies, Netflix, and other services on devices whose displays support the standard. A new media player and additional media metric infused in certain classes are now also part of the package, as are improvements to the MediaDRM class meant to make the anti-piracy solution more effective.
The JobScheduler tool should be better at managing network loads on Android P, whereas the Neural Networks API from Android 8.1 Oreo now supports nine new operations, including a “Squeeze” function that appears to be solely targeting pressure-sensitive frames such as the ones found inside the HTC U11, Pixel 2, and the Pixel 2 XL. Oreo’s autofill framework is also getting more versatile and may soon be fully embraced by Google itself, with some of its popular apps like Chrome still using the legacy solution relying on Android’s Accessibility Services over half a year after Oreo hit the stable channel. Support for system backup encryption tied to a client-side secret is included in Android P, as is a unified fingerprint authentication dialog and a new prompt meant to verify user intent in regards to completing highly sensitive transactions. Tracking individual app’s windows in scenarios when multiple redraws are occurring should also be easier in Android P, courtesy of one newly included API which provides developers with a native service for detecting new titles in panes, disappearing panes, and other such phenomena that were previously much harder to track.
Google’s first experimental build of Android P includes several “Convenience Actions” meant to serve as development tools for creating apps that can take screenshots and lock devices without resorting to system workarounds, as well as manage tooltip text prompts. Many new Accessibility services have also been included, with the new OS being capable of providing more detailed information about its section titles, in addition to coming with support for more streamlined heading-based and group navigation actions. The first developer preview of Android P can presently be manually flashed onto the Pixel and Pixel 2-series smartphones; developers and enthusiasts interested in giving it a go can obtain all of the necessary files to do so by referring to the banner below. Google is likely to debut the second preview build of the mobile OS at its annual I/O developer conference which is scheduled to start on May 8 in its hometown of Mountain View, California.