Acation bar in New Android 3.0

Acation bar in New Android 3.0

Android version 3.0 introduced a standardized navigation element called the ActionBar. This bar sits at the top of the screen and contains common actions and navigation elements, such as tabs. It also includes an icon on the left side that can be used as an up button (see Figure 1).

Figure 1 The up button as displayed in the Google Play app.

Difference Between UP and BACK?

It's important to understand the differences between the up and the back button. First, the up button is only relevant within your app. It will never take the user to a different application. In addition, the concept of up means that the user will return to a "logical" parent of the current activity. In the Gmail app, for example, pressing the up button while viewing an email message would take you to a list of email messages. If no parent exists, up should take the user to the "home" activity of your app.

The back button, on the other hand, moves back through the stack of recent activities. This is limited to those activities in the current task (more about that in a moment). Think of it like the browser's back button. As a user navigates through an app and into others, the back button will take him backward through those activities, until he reaches the launcher. Once the user reaches the launcher, he can no longer press back.

How Tasks Works and Relate to the Back Stack?

Tasks are collections of activities opened in chronological order by the user. A new task is created when the user opens an application from the launcher. Moving from one application to another generally doesn't start a new task. Instead, the new activity becomes part of the existing task. For example, if you navigate from an existing app to the Google Maps app, the displayed map will be part of the existing app task. Pressing back from the map will take the user back to the original app.
Each Android task has an associated back stack. The back stack can seem complicated at first, but its implementation is firmly rooted in basic computer science. It's just a stack. The elements in the stack are the screens of an app, called activities. As you navigate from one activity to another, the previous activity is pushed onto the stack. When you press the back button, the current activity (the current screen displayed to the user) is destroyed, and the activity at the top of the back stack is popped off and displayed to the user. If the back stack is empty, the user is taken back to the launcher.
An important point to note is that the system will not keep tasks around forever. When it needs resources, it will often clear the task and corresponding back stack. This is quite common, and you shouldn't rely on any particular back stack state to exist for your app to function correctly. If you really need it, however, it's possible to disable this behavior by setting the alwaysRetainTaskState attribute on an activity in your manifest.