SEG 3210 User Interface Design

1
SEG 3210User Interface Design Implementation

• Wael Hassan
• SEG3120 Unit-C Part I

2
Unit C Task Analysis

• Introduction to Task Analysis
• Tasks and Goals Revisited
• Approaches to task analysis
• A Procedure for task analysis
• Case Study File Manipulation on Command Line
  vs. GUI
• Mail-Order Case Study Classes of User

3
1. Introduction to Task Analysis

• Definitions of task analysis
• The process of analyzing and improving the way
  people perform their jobs
• What people do ? The things they do
• What things they work with ? The things they act
  on
• What they must know ? The things they need to know

• Provides an essential basis for
• Early stages of interface design
• Evaluation of design at early and late stages
• Structuring training in system use
• Structuring and design of documentation

4
Example TASK Clean the house

• In order to clean the house you need to
• Get the vacuum cleaner out
• Fix the appropriate attachments
• Clean the rooms
• When the dust bag gets full, empty it
• Put the vacuum cleaner and tools away
• You must know about
• Vacuum cleaners, their attachments, dust bags
• Cupboards in which the vacuum cleaner is kept
• rooms to be cleaned, etc.

5
2. Tasks and Goals Revisited

• Goals and tasks are sometimes equated
• We will separate them
• What is a task?
• A procedure
• Performed by one or more classes of user
• Directed towards a goal
• May have a super-task
• Broken down into subtasks
• With possible loops and alternation (if-then)
• Involves actions (at the bottom level)
• A task that involves no problem solving and
  cannot be usefully broken down into components
• ? Avoid thinking of a task in terms of what the
  computer alone must do

6
2. Tasks and Goals Revisited

• What is a goal?
• A state the user wants to be in
• Achieved by performing a task
• Sub-goals are the goals of subtasks
• The user has to figure out what task to perform
• May or may not be explicitly known in advance of
  task execution
• e.g. Implicit goal Writing a letter
• general task is known
• details evolve
• goal is achieved when user is satisfied
• e.g. Explicit goal Adding an item to inventory
• task and goal are known by an experienced user
• ? High level goals are less likely to be explicit
• Other perspectives
• A goal is often equated with its task

7
3. Approaches to task analysis

• Task decomposition
• Splitting task into (ordered) subtasks
• The order in which these are performed
• Knowledge based techniques
• What the user knows about the objects and actions
  involved in a task and how that knowledge is
  organised
• Focus on
• Objects - used in task
• Actions - performed
• Taxonomies represent levels of abstraction
• Entity-relation based analysis
• Relationships between objects, actions and the
  people who perform them
• General method
• Observe
• Collect unstructured lists of words and actions
• Organize using notation or diagrams

8
4. A Procedure for Task Analysis

• (case studies are found in subsequent sections)
• Two basic steps
• Model tasks
• Simplify and improve each task model
• Use the simplified task model as a basis for the
  user interface.

9
Model tasks

• (easier if there is an existing system)
• Observe and interview users
• and/or brainstorm
• and/or prototype
• and/or study documentation
• Determine the classes of users
• Manager, salesperson, shipper, client, etc.
• b) Create a list of all tasks to be performed by
  users
• c) Prioritize the tasks by frequency of use and
  importance
• Later on, perform more detailed analysis on
  higher-priority tasks
• If there is an existing system, one can measure
  the usage of commands
• ? Note different classes of user will use
  tasks with different frequencies

10
Model tasks

• d) Gather other detailed information about each
  task
• Why? Determine the goals
• Determine preconditions
• ... what users need to know to perform the task
• ... what are the data and objects involved
  (perhaps from a parallel object oriented
  analysis)
• How? Decompose the task into subtasks and actions
• Determine side-effects of performing the task
• things changed that are not directly relevant to
  the goal state
• Determine users mental models
• there may be several
• Determine any interrelationships between tasks
• Determine possible breakdowns
• things that can go wrong
• Iteratively perform this step for subtasks
• e) Diagram each task
• Possible techniques
• hierarchies of subtasks (probably best)
• time-lines (scenario diagrams)
• flow-charts

11
Simplify and improve each task model

• a) Abstract task models as much as possible
• Determine what the real top level goals are
• Remove details (lower level subtasks) that can
  be changed
• i.e. subtasks that are part of this office
  environment or user interface
• situations where different systems would have
  somewhat different subtasks
• b) Build or improve conceptual models (discussed
  later) to help simplify the users thinking about
  tasks
• c) Add back detailed layers one-by-one
• Think of several alternative sets of subtasks and
  determine which is better

12
Simplify and improve each task model

• c) Add back detailed layers one-by-one
• Use the following guidelines
• Standardize subtasks
• so similar subtasks are performed in similar ways
• Rearrange the order of subtasks to make them
  easier to perform
• Reduce the need for the user to perform subtasks
  (i.e. reduce overhead)
• Minimize setup and configuration by using
  defaults
• Reduce the need for users to recall from memory
• Where possible have the computer calculate
  whatever it can
• Eliminate subtasks if their benefit is less than
  their cost
• Ensure the system guides the user through the
  tasks
• The system should know how the user works
• It should help the user set up goals
• Account for the possibility of multitasking
• If a user jumps around among tasks ensure steps
  need not be repeated and that the user can pick
  up where she left off

13
5. Case Study File Manipulation on Command Line
vs. GUI

• Goal
• Execute a command
• on a command line that has one or two filename
  arguments
• To be performed
• frequently
• by
• command-line users (often experts these days)
• Preconditions
• Command verb is known
• Filenames are known
• Decomposition
• Ensure command prompt is displayed
• Type command verb
• Enter first filename
• Decide If no second filename, go to step 6
• Enter second filename
• Verify command
• Press ltreturngt

14
5. Case Study File Manipulation on Command Line
vs, GUI

• Breakdowns
• Command verb incorrect
• Filenames(s) incorrect
• Wrong number of arguments specified
• Arguments in wrong order
• Operation not allowed on filenames

15
Goal Delete a file in DOS

• To be performed
• frequently
• by
• command-line users
• Preconditions
• File to delete is known
• System is displaying the command prompt
• Decomposition
• Retrieve the command verb for deleting a
  file,e.g. ERASE or DELETE
• Think of directory name and filename
• Enter the command

16
Goal Delete a file in DOS

• Analysis to improve task model (Possible real
  top-level goals)
• Make space on disk for something else
• Possible simplification Automatic compressing /
  archiving of old files
• Get rid of unwanted clutter
• Possible simplification Build version control
  into file system so multiple versions do not
  accumulate
• Ensure nobody else can read file
• Other possible simplifications
• Eliminate need for recall and typing in step 1 by
  providing a menu
• Simplify steps 2 and 3 by providing direct
  manipulation
• Have the system prompt the user if the command is
  entered without the file spec

17
Goal Drag an item to its destination in a GUI
desktop

• To be performed frequently by all users
• Preconditions
• Item to drag is known
• Effect of dropping item on destination is known
• Decomposition
• Locate icon or name of item on screen
• Move cursor to icon
• Press and hold mouse button
• Locate destination on screen
• Move cursor to destination
• Verify that destination highlighted
• Release mouse button
• Breakdowns
• Item to drag is not draggable
• Cursor not over icon in step 2
• Mouse released early or not over destination
• Destination invalid for item being dragged

18
Goal Delete a file in a GUI desktop

• To be performed frequently by all users
• Preconditions
• File to delete is known
• Decomposition
• Bring file manager to the foreground(how to do
  this depends on the OS version)
• Uncover trash/recycle icon if hidden
• Drag file to the trash/recycle bin
• Users mental model
• File will be sitting in the trash/recycle bin and
  can be retrieved
• Possible simplifications
• Provide a keyboard shortcut for step 1
• Provide a keyboard shortcut for steps 2 and 3

19
6. Mail-Order Case Study Classes of User

• Client rep/salesperson
• Takes phone orders and queries
• Needs to be able to work very fast
• Client
• Dials in to place orders/queries
• Needs very simple approach
• Shipper
• Assembles orders to be sent to clients when
  sufficient inventory is present
• Puts items in crates and sends them to clients
• Records that items have been sent
• Return handler
• Handles returns of merchandise by client

20
6. Mail-Order Case Study Classes of User

• Supplier orderer
• Sends out orders printed by the system in
  response to client orders and/or low inventory
• Receiver
• Matches packages received from suppliers to
  supplier orders
• Updates inventory
• Inventory taker
• Verifies inventory matches system
• Adjusts system as necessary, reporting
  discrepancies.
• Backorder processor
• Sends polite letters when delays occur
• Pursues suppliers so orders can be fulfilled
• Accounts receivable clerk
• Updates system when accounts have been paid
• Pursues derelict clients
• Sets credit limits
• Inventory management analyst
• Decides how much of each item should be kept in
  inventory
• Manager
• Performs general queries