User Analysis

Services like AppleTV and Netflix augment the TV viewing experience. Consequently, these services aim to be usable for as large a fraction of TV viewers as possible. We believe that users of game consoles are generally a subset of TV viewers, so we do not identify a separate user group for this application.

Netflix/AppleTV/Game console users span every age group and nearly all backgrounds. We believe the following are true about our target users:

• They can use a TV remote/game controller. Several first time Xbox users we observed were able to very quickly learn how to use the controller. This included complete beginners as well as users who had used other gaming consoles.
• Although most people can use game controllers, we noticed that experienced gamers were much faster at manipulating the controller than less experienced users.
• They can read an on screen English keyboard like the ones used by Netflix and Xbox.
• They typically use these services sitting in their own home, often reclining on couches and displayed on very large screens.
• They are already familiar with how to operate the TV and any other relevant device (like the Xbox).
• They currently use or at least know that there is a search feature.

Rewrite Additions:
For the purpose of this project, the user group will be specified to college students.  This will set up an age group of about 18-22 with at least a high school education.  This population knows how to use devices which can be used to watch movies, such as an AppleTV, XBox, or BluRay player.  However, we will be making our product simple enough to extend usability to other user groups that might not have as much experience with current-day televisions and gaming devices.

Task Analysis

Why is the task being done?
    The user needs to input character sequences. eg to search for show/movie, enter his username , or send messages, etc.

What does the user need to know or have before doing the task?
    He needs to know how to get to the text entry field and how to insert/delete characters into it.

Where is the task performed?
    The user is watching TV or playing a video game, so he is likely at home and sitting down.

What is the environment like?
    If the user is alone, it is likely quiet and relaxed so it should be easy for him to focus. If the user is with others, it may be loud and distracting to the user.

How often is the task performed?
    Text entry is relatively infrequent but quite an important part of these applications. If the user depends on search to find what he is looking for, the text entry field is on the application's critical path and significantly contributes to the user's overall satisfaction. Xbox users we observed often sent messages to each other.

What are its time/resource constraints?
    Text entry is limited by (1) the size of the screen on which text entry widgets can be displayed (2) the controls available on the remote/controller (3) the user's patience (it shouldn't take more than a few seconds to type a few words.

How is the task learned?
    The user might discover how to enter text by exploring the interface (i.e. pressing buttons on the control and seeing what happens), or he might be guided by on screen instructions (a message: "use the arrow keys to move", etc.), or he might learn by watching others and/or receiving instruction.

What can go wrong?
    The remote might not be pointed in the right direction, the batteries may die, the user may select the wrong letter.

Who else is involved in the task?
    Other people may tell the user what the type, but ultimately only one person is performing the text entry.

Rewrite Additions:
The specific user task we will be supporting will be having a user search for a movie on an internet video application, such as Netflix, Hulu, or AmazonMovies.  The user will need to be able to sign in to the application, which means that one task will be to enter a username.  For this there will be no autocomplete because people might not necessarily choose regular names for usernames, but the user can see what he is typing.  The next task will be to type in a password.  This will need to be hidden, so we will need to decide a way to hide the characters being chosen.  The last task will be to search for a movie to watch.  The user will begin typing letters, and an autocomplete will show the top movies based on the current typed text.

High Level Tasks

• What should the on screen widgets look like and do?
  • Current systems use a rectangular grid of uppercase characters listed in alphabetical order. We need to design another widget that is faster and easily navigable with a remote/controller. Potential ideas include cascading lists of characters, circular rings of highly probably letters, or something like Dasher.
• What should the controls on the remote do to the widgets?
  • Currently, arrow keys are used to change which letter in the rectangular grid is selected, and another key is pressed to actually select that letter. A special on-screen key should be selected to backspace or move the cursor between the already selected letters which are displayed in a box. We might like to take advantage of more keys on the remote to more quickly manipulate the widget and reach the desired character faster.
• How should errors be conveyed and corrected by user?
  • The user may press and arrow key several times and because some devices may be slow to process rapid keystrokes, errors may occur regarding which letters have been seleted. There should be some immediate feedback to let the user know that their keystrokes are still being processed. Feedback could take the form of an on-screen indicator or vibrating the controller. Other errors may be entirely user generated, such as selecting the wrong letter. If we can detect that the user likely meant another letter, we should have a non-intrusive yet accessible way of allowing the user to select the correct version.