FIT3084: Introduction
to Interactivity
In the previous lecture:
About Human-Computer Interaction (HCI) and design issues for Multimedia.
In this lecture:
Every aspect of interactive software must be designed and
tested for/by the human target audience.
Not just by the software engineers and programmers who wrote
the thing!
In the case of websites here are some important issues...
Elements must provide an appropriate mood and operate within a metaphor
aimed at the target audience.
A simple rule states... if a user will perform a task
|
For example, if it took three keystrokes to save a word processor file every time, a task which ought to be performed often, a user would rapidly become frustrated. A single key chord (group of keys held down together) such as ALT-S may easily achieve the same task.
The act of setting preferences on the other hand, a task done
infrequently, may require the user to navigate several menus and dialogue boxes.
This is OK because the user usually only sets preferences for the software every
now and again.
Users perform an interaction task or several interaction tasks
in sequence an interaction dialogue.
An interaction task is the entry of a unit of information using a hardware
or software device.
Four basic interaction tasks:
|
Interaction techniques are the means of achieving interaction tasks using input devices.
E.g. A selection task may be achieved:
A single input device may be used for more than one kind of interaction task.
E.g. A mouse may be used for:
The most common devices for interaction with computer based multimedia are:
What is the dominant interaction task on the Web? Which hardware device is used
the most frequently?
Locator devices (for positioning) may be conveniently categorized
Absolute locators:
|
|
Relative locators:
|
(Common relative locators: mice; trackball)
Direct locators:
user points directly at the screen.
Light
Pen (Pen tip and cursor at the same spatial location) |
Touch
Screen (Finger tip and cursor at the same spatial location) |
Indirect locators:
user moves a cursor across the screen using a device separate from the screen.
E.g. mouse, joystick
Joystick (Hand on joystick and rocket ship on screen at different spatial locations) |
Discrete locators: require repeated operation
to achieve a movement in small increments in a given direction. |
(Cursor control keys) |
Continuous locators: require a single smooth operation
to achieve a movement in a given direction.
E.g. joystick, mouse, graphics tablet
The ratio of the speed/distance of hand movement (Control) to the speed/distance of cursor movement (Display) for a continuous device.
Large ratio - large hand movement : small cursor movement
Good for accurate positioning, poor for long movements.
Small ratio - small hand movement : large curosor movement
Good for rapid movement across long distances, poor for accurate positioning.
A relative device may be set up to change its C/D ratio depending
on the velocity of movement.
Should the ratio be high or low for high-velocity movements? Why?
Linguistic (specify in some coordinate system)
Boxes to quantify position, orientation and scale of an object.
From the UI of Infini-D, a 3D animation package.
Spatial (move cursor to position)
Buttons to change position and orientation of an object by selection.
From the UI of Infini-D, a 3D animation package.
Device movement direction should correspond to movement in screen space. For
instance, if the mouse moves left, so should the cursor or selected object!
Key spatial orientation for cursor control should correspond to onscreen
movement. |
Grids may aid spatial positioning tasks by 'snapping' the cursor
to specific points in a plane, and not allowing the cursor to locate objects
(for example line endpoints in a drawing package) at non-grid points.
|
Selection involves choosing an element from a choice set.
Choice sets are either predominantly fixed, or variable.
Choice sets include:
Fixed and variable choice sets may call for different methods of selection...
Menus for selection from fixed sized choice sets may be:
Menus of commands may be grouped functionally and displayed textually,
Pointing (or naming with auto-completion) for selection from variable
sized choice sets.
Naming is preferable where there is insufficient screen space to display all
objects for pointing. However, this method assumes some knowledge from the user!
(What is it?)
Tips for onscreen location (for mouse selection) - note the contradictory demands
What are the easiest places for a user to position the cursor using a mouse?
Why is this so? How can you use this knowledge to build an effective interface?
Vital Principles of Interface Design
|
©Copyright Alan Dorin 2008