EECS 578 Week 1

Introduction and the Design of Everyday Things


Information about the Course


Warning about jargon


What is this Course About

This is a course in sensitivity.

Making a program work for you is pretty easy. Making it work for another user is much harder  You need to think about the actual users of your software early in the design stage, and keep their needs in mind all the time.

Engineers are not trained in design and psychology and these are increasingly important in writing good software. People in general have a hard time sitting quietly while people criticize your hard work, but this is very important to creating a useful product.

This course is not about using widgets to create Windows/Icon based interfaces. It is about the procedures to go through to create good designs and then evaluate them.

First we will look at good and bad design in general, independent of computer programs. We will look at how to effectively present information. We will look at tradional user interfaces such as keyboards and mice. We will look at how to perform evaluation of these traditional interfaces and look at specific case studies. We will finally look at interfaces to support groupwork and more current human-computer interfaces such as speach, immersive environments.


The Design of Everyday Things

(material from: The Design of Everyday Things by Donald Norman)


Science Finds
Industry Applies
Man Conforms
(1933 chicago World's Fair)

Start by looking at Interfaces for

"Affordances - perceived and actual properties of a thing, primarily those fundamental properties that determine just how the thing could possibly be used." Affordances provide strong clues to the operation of a thing Provide a good conceptual model and make things visible


Two concptual models

the designer's conceptual model
the user's conceptual model.

The system image is the visible part of a device (including the physical structure, the documentation, instructions, etc). The designer only talks to the user through the system image. If the system image doesn't make the design model clear then the user will create a differnet model through their interaction.

Mental models - conceptual model of the way something works, often constructed from fragmentary evidence


Thermostat - will a room or oven heat (or cool) faster if the thermostat is turned all the way to the maximum setting?

    2 'folk' theories of thermostats
        timer theory - thermostat controls the relative proportion of time that the device stays on
        valve theory - thermostat controls how much heat (cold) comes out of the device

    both are wrong
        thermostat is an on/off switch - fully on or fully off - no inbetween

    design gives no hint to the actual model
    user's form their own theories


Why is the basic automobile easy to figure out?


Mapping - relationship between controls and their affects. Want the mapping to be 'natural' - taking advantage of physical analogies and cultural standards.
    steering wheel
    stove burners
    light switches

    Examples from my new house
        no writeups on the appliances
        trial and error

        light switch for fan/lights in family room
        controls for the stove
        dryer


Constraints

Forcing function
    key in car and the starter now the same - in past it was possible to try and start car without
        the key in the ignition
    in some cars you need to use the keys to lock the dor - cant lock keys inside

    interlocks - prevent accidental use of a device
        force operations to take place in proper sequence
        pin on fire extinguisher or hand grenade
    lockins - keeps operation active preventing someone from permanently stopping it
        word processor
    lockouts - prevents an event from occuring
        in case of fire, cant go below the first floor on stairs


7 stages of action

    eg    reading a book at dusk and its getting dark.
    goal - get more light
    forming intention - push the switch on the lamp
    specifying action - how to move body

7 Stages as Design Aid

How easily can a person
    Determine the function of the device
    Tell what actions are possible
    Determine mapping from  intention to physical movement
    Perform the action
    Tell if system is in desired state
    Determine mapping from system state to interpretation
    Tell what state system is in


Gulf of execution
    how well does the system allow someone to do their intended actions directly
    do the affordances provided by the system match the actions intended by the person
    bad if not clear what actions need to be done to accomplish the intention

Gulf of evluation
    how well does the system provide a visible state that can be directly perceived and that
        is interperable in terms of the intentions and expectations of the user
    how much effort user must exert to interpret the state of the system and determine  how
        well the expectations and intentions have been met


Designing for error

1. Understand the cause of error and design to minimize those causes
2. Make it possible to reverse (undo) actions, or make it harder to do what cannot be reversed
3. Make it easier to discover errors that do occur and make them easier to correct
4. Think of an object's user as attempting to do a task, getting there by imperfect approximations, Don't think of the user as making errors, think of the actions as approximations of what is desired
 


Visibility - User can tell the state of the device and the alternatives for action
Good conceptual model - User given consistent in presentation of operations and results
Good Mappings - easy to determine relationships between actions and results, controls and
    their effects, system state and whats visible
Feedback - User receives full and continuous feedback


Coming Next Time

Information Visualization

e.g. the following figure described as possibly the best statistical graph ever drawn


from p41 of TheVisual Display of Quantitative Information

Illustration drawn by Charles Joseph Minard in 1861 portraying Napoleon's losses during his 1812 march to and from Moscow

The chart show 6 variables
    size of the army
    location on a 2D surface
    direction of the army's movement
    temperature
 



Department of Electrical Engineering and Computer Science
University of Illinois at Chicago