A
INTER, ,CTIOW
DESIGN
I
beyond human-computer interaction
Color Plate 1
Figure 1.2 Novel forms of interactive products embedded with computational power (clockwise from top left):
(i) Electrolux screenfridge that provides a
range of functionality, including food management where recipes are
displayed, based on the
food stored in the fridge.
[IV)Barney, an interactive cuddly
toy that makes learning enjoyable.
(iii) 'geek chic', a Levi jacket equipped
with a fully integrated computer network
(body area network), enabling the wearer
to be fully connected to the web.
ENTER
Figure 1.1 1 2D and 3D buttons. Which are easier to distinguish between?
Color Plate 2
Figure 2.1 An example of augmented reality. Virtual and
physical worlds have been combined so that a digital image of
the brain is superimposed on the person's head, providing a
new form of medical visualization.
Figure 2.14 The i-room project at Stanford: a graphical
rendering of the Interactive Room Terry Winograd's
group is researching, which is an innovative technologyrich prototype workspace, integrating a variety of displays and devices. An overarching aim is to explore new
possibilities for people to work together (see
http://graphics.stanford.EDU/projects/iwork/).
.
I..
-
,
.
Color Plate 3
Figure 2.6 Recent direct-manipulation virtual environments
(a) Virtue (Daniel Reid, 1999, www-pablo.cs.uiuc.edulProjectNRNirtue) enables software developers to directly manipulate software components and their behavior.
(b), (c) Crayoland (Dave Pape, www.ncsa.uiuc.eduNis/) is an interactive virtual environment where the child
in the image on the right uses a joystick to navigate through the space. The child is interacting with an avatar in
the flower world.
Color Plate 4
Figure 3.7 Dynalinking used in the PondWorld software. In the background is a simulation
of a pond ecosystem, comprising perch, stickleback, beetles, tadpoles, and weeds. In the
foreground is a food web diagram representing the same ecosystem but at a more abstract
level. The two are dynalinked: changes made to one representation are reflected in the
other. Here the user has clicked on the arrow between the tadpole and the weed represented in the diagram. This is shown in the PondWorld simulation as the tadpole eating
the weed. The dynalinking is accompanied by a narrative explaining what is happening and
sounds of dying organisms.
Figure 3.9 A see-through
handset-transparency does not
mean simply showing the insides of
a machine but involves providing a
good system image.
Color Plate 5
Figure 4.1 'l'he rooftop gar-
den in BowieWorld, a collaborative virtual environment
(CVE) supported by
Worlds.com. The User takes
part by "dressing up" as an
avatar. There are hundreds of
avatars to choose from, including penguins and real
people. Once avatars have
entered a world, they can explore it and chat with other
avatars.
Color Plate 6
Figure 5.3 Examples of aesthetically pleasing interactive products: iMac, Nokia cell phone
and IDEO's digital radio for the BBC.
1
Figure 5.9 Virtual screen characters:
(a) Aibo, the interactive dog.
Color Plate 7
Figure 5.1 1
I-lerman the bug
watches as a student chooses
roots for a plant
in a n Alpinc
meadow.
Figure 5.1 2 The
Woggles interface, with icons
and slider bars
repl-escnting
emotions. specch
and actions.
Color Plate 8
Figure 7.3(b) The KordGrip being used underwater
Figure 5.13 Rea the real estate
agent welcoming the user to look
at a condo.
Figure 15.8 The first foam models of a mobile communicator for
children.
INTERACTION'
DESIGN
beyond human-computer interaction
John Wiley & Sons, Inc.
ACQUISITIONS EDITOR
MARKETING MANAGER
SENIOR PRODUCTION EDITOR
COVER DESIGNER
ILLUSTRATION EDITOR
ILLUSTRATIONS
COVER IMAGE
Gaynor Redvers-MuttonlPaulCrockett
Katherine Hepburn
Ken Santor
Madelyn Lesure
Anna Melhorn
Tech-Graphics, Inc.
"Thoughts in Passage 11" by Michael Jon March.
Courtesy of Grand Image Publishing
This book was set in 10112 Times Ten by UG I GGS Information Services, Inc., and printed
and bound by R. R. DonnelleylCrawfordsville.The cover and the color insert were printed
by Phoenix Color Corporation.
This book is printed on acid free paper. m
Copyright O 2002 John Wiley & Sons, Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any
form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise,
except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without
either the prior written permission of the Publisher, or authorization through payment of the
appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA
01923, (508) 750-8400, fax (508) 750-4470. Requests to the Publisher for permission should be
addressed to the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue, New
York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008, E-Mail:
PERMREQ@WILEY.COM.
To order books or for customer service please call 1(800)225-5945.
Library of Congress Cataloging in Publication Data.
Preece, Jennifer.
Interaction design : beyond human- computer interaction1Jennifer Preece, Yvonne Rogers, Helen
Sharp.
p. cm.
Includes bibliographicalreferences and index.
ISBN 0-471-49278-7 (paper : alk. paper)
1. Human-computer interaction. I. Rogers, Yvonne. 11. Sharp, Helen. 111. Title.
QA76.9.H85 P72 2002
004'.01'94c21
Printed in the United States of America
2001006730
Preface
Welcome to Interaction Design: Beyond Human-Computer Interaction, and our interactive website at ID-Book.com
This textbook is for undergraduate and masters students from a range of backgrounds studying classes in human-computer interaction, interaction design, web
design, etc. A broad range of professionals and technology users will also find this
book useful, and so will graduate students who are moving into this area from related disciplines.
Our book is called Interaction Design: Beyond Human-Computer Interaction
because it is concerned with a broader scope of issues, topics, and paradigms than
has traditionally been the scope of human-computer interaction (HCI). This reflects
the exciting times we are living in, when there has never been a greater need for interaction designers and usability engineers to develop current and next-generation
interactive technologies. To be successful they will need a mixed set of skills from
psychology, human-computer interaction, web design, computer science, information systems, marketing, entertainment, and business.
What exactly do we mean by interaction design? In essence, we define interaction design as:
"designing interactive products to support people in their everyday and working lives".
This entails creating user experiences that enhance and extend the way people
work, communicate, and interact. Now that it is widely accepted that HCI has
moved beyond designing computer systems for one user sitting in front of one machine to embrace new paradigms, we, likewise, have covered a wider range of issues. These include ubiquitous computing and pervasive computing that make use
of wireless and collaborative technologies. We also have tried to make the book
up-to-date with many examples from contemporary research.
The book has 15 chapters and includes discussion of how cognitive, social, and
affective issues apply to interaction design. A central theme is that design and evaluation are interleaving, highly iterative processes, with some roots in theory but
which rely strongly on good practice to create usable products. The book has a
'hands-on' orientation and explains how to carry out a variety of techniques. It also
has a strong pedagogical design and includes many activities (with detailed comments), assignments, and the special pedagogic features discussed below.
The style of writing is intended to be accessible to students, as well as professionals and general readers, so it is conversational and includes anecdotes, cartoons, and case studies. Many of the examples are intended to relate to readers'
own experiences. The book and the associated website encourage readers to be active when reading and to think about seminal issues. For example, one feature we
have included in the book is the "dilemma," where a controversial topic is aired.
The aim is for readers to understand that much of interaction design needs consid-
vi
Preface
eration of the issues, and that they need to learn to weigh-up the pros and cons and
be prepared to make trade-offs. We particularly want readers to realize that there
is rarely a right or wrong answer although there are good designs and poor designs.
This book is accompanied by a website, which provides a variety of resources
and interactivities, The website offers a place where readers can learn how to design
websites and other kinds of multimedia interfaces. Rather than just provide a list of
guidelines and design principles, we have developed various interactivities, including online tutorials and step-by-step exercises, intended to support learning by
doing.
Special features
We use both the textbook and the web to teach about interaction design. To promote good pedagogical practice we include the following features:
Chapter design
Each chapter is designed to motivate and support learning:
Aims are provided so that readers develop an accurate model of what to expect in the chapter.
Key points at the end of the chapter summarize what is important.
Activities are included throughout the book and are considered an essential
ingredient for learning. They encourage readers to extend and apply their
knowledge. Comments are offered directly after the activities, because pedagogic research suggests that turning to the back of the text annoys readers
and discourages learning.
An assignment is provided at the end of each chapter. This can be set as a
group or individual project. The aim is for students to put into practice and
consolidate knowledge and skills either from the chapter that they have just
studied or from several chapters. Some of the assignments build on each
other and involve developing and evaluating designs or actual products.
Hints and guidance are provided on the website.
Boxes provide additional and highlighted information for readers to reflect
upon in more depth.
Dilemmas offer honest and thought-provoking coverage of controversial or
problematic issues.
Further reading suggestions are provided at the end of each chapter. These
refer to seminal work in the field, interesting additional material, or work
that has been heavily drawn upon in the text.
Interviews with nine practitioners and visionaries in the field enable readers
to gain a personal perspective of the interviewees' work, their philosophies,
their ideas about what is important, and their contributions to the field.
Cartoons are included to make the book enjoyable.
How to use this book
vii
ID-Book.com website
The aim of the website is to provide you with an opportunity to learn about interaction design in ways that go "beyond the book." Additional in-depth material,
hands-on interactivities, a student's corner and informal tutorials will be provided.
Specific features planned include:
Hands-on interactivities, including designing a questionnaire, customizing a
set of heuristics, doing a usability analysis on 'real' data, and interactive tools
to support physical design.
Recent case studies.
Student's corner where you will be able to send in your designs, thoughts,
written articles which, if suitable, will be posted on the site at specified times
during the year.
Hints and guidance on the assignments outlined in the book.
Suggestions for additional material to be used in seminars, lab classes, and
lectures.
Key terms and concepts (with links to where to find out more about them).
Readership
This book will be useful to a wide range of readers with different needs and
aspirations.
Students from Computer Science, Software Engineering, Information Systems,
Psychology, Sociology, and related disciplines studying courses in Interaction Design and Human-Computer Interaction will learn the knowledge, skills, and techniques for designing and evaluating state-of-the-art products, and websites, as well
as traditional computer systems.
Web and Interaction Designers, and Usability Professionals will find plenty to
satisfy their need for immediate answers to problems as well as for building skills to
satisfy the demands of today's fast moving technical market.
Users, who want to understand why certain products can be used with ease
while others are unpredictable and frustrating, will take pleasure in discovering
that there is a discipline with practices that produce usable systems.
Researchers and developers who are interested in exploiting the potential of the
web, wireless, and collaborative technologies will find that, as well as offering guidance, techniques, and much food for thought, a special effort has been made to include examples of state-of-the-art systems.
In the next section we recommend various routes through the text for different
kinds of readers.
How to use this book
Interaction Design is not a linear design process but is essentially iterative and
some readers and experienced instructors will want tb find their own way through
the chapters. Others, and particularly those with less experience, may prefer to
viii
Preface
work through chapter by chapter. Readers will also have different needs. For example, students in Psychology will come with different background knowledge and
needs from those in Computer Science. Similarly, professionals wanting to learn
the fundamentals in a one-week course have different needs. This book and the
website are designed for using in various ways. The following suggestions are provided to help you decide which way is best for you.
From beginning to end
There are fifteen chapters so students can study one chapter per week during a
fifteen-week semester course. Chapter 15 contains design and evaluation case studies.
Our intention is that these case studies help to draw together the contents of the
rest of the book by showing how design and evaluation are done in the real world.
However, some readers may prefer to dip into them along the way.
Getting a quick overview
For those who want to get a quick overview or just the essence of the book, we
suggest you read Chapters 1, 6, and 10. These chapters are recommended for
everyone.
Suggestions for computer science students
In addition to reading Chapters 1,6, and 10, Chapters 7 and 8 contain the material
that will feel most familiar to any students who have been introduced to software
development. These chapters cover the process of interaction design and the activities it involves, including establishing requirements, conceptual design, and physical design. The book itself does not include any coding exercises, but the website
will provide tools and widgets with which to interact.
For those following the ACM-IEEE Curriculum (2001)*, you will find that this
text and website cover most of this curriculum. The topics listed under each of the
following headings are discussed in the chapters shown:
HC1 Foundations of Human-Computer Interaction (Chapters 1-5, 14,
website).
HC2 Building a simple graphical user interface (Chapters 1,6,8,10 and the
website).
HC3 Human-Centered Software Evaluation (Chapters 1,10-15, website).
HC4 Human-Centered Software Design (Chapters 1,6-9,15).
HC5 Graphical User-Interface Design (Chapters 2 and 8 and the website.
Many relevant examples are discussed in Chapters 1-5 integrated with discussion of cognitive and social issues).
*ACM-IEEE Curriculum (2001) [computer.org/education/cc2001/]is under development at the time of
writing this book.
How to use this book
ix
HC6 Graphical User-Interface Programming (touched upon only in Chapters 7-9 and on the website).
HC7 HCI Aspects of Multimedia Information Systems and the web (integrated into the discussion of Chapters 1-5, and in examples throughout the
text, and on the website).
HC8 HCI Aspects of Group Collaboration and Communication Technology
(discussed in 1-5, particularly in Chapter 4. Chapters 6-15 discuss design and
evaluation and some examples cover these systems, as does the website.)
Suggestions for information systems students
Information systems students will benefit from reading the whole text, but instructors
may want to find additional examples of their own to illustrate how issues apply to
business applications. Some students may be tempted to skip Chapters 3-5 but we recommend that they should read these chapters since they provide important foundational material. This book does not cover how to develop business cases or marketing.
Suggestions for psychology and cognitive science students
Chapters 3-5 cover how theory and research findings have been applied to interaction design. They discuss the relevant issues and provide a wide range of studies
and systems that have been informed by cognitive, social, and affective issues.
Chapters 1 and 2 also cover important conceptual knowledge, necessary for having
a good grounding in interaction design.
Practitioner and short course route
Many people want the equivalent of a short intensive 2-5 day course. The best
route for them is to read Chapters 1,6,10 and 11 and dip into the rest of the book
for reference. For those who want practical skills, we recommend Chapter 8.
Plan your own path
For people who do not want to follow the "beginning-to-end" approach or the suggestions above, there are many ways to use the text. Chapters 1,6,10 and 11 provide
a good overview of the topic. Chapter 1 is an introduction to key issues in the discipline and Chapters 6 and 10 offer introductions to design and evaluation. Then go
to Chapters 2-5 for user issues, then on to the other design chapters, 2-9, dipping
into the evaluation chapters 10-14 and the case studies in 15. Another approach is to
start with one or two of the evaluation chapters after first reading Chapters 1, 6, 10
and 11, then move into the design section, drawing on Chapters 2-5 as necessary.
Web designer route
Web designers who have a background in technology and want to learn how to design usable and effective websites are advised to read Chapters 1, 7, 8, 13 and 14.
x
Preface
These chapters cover key issues that are important when designing and evaluating
the usability of websites. A worked assignment runs through these chapters.
Usability professionals' route
Usability professionals who want to extend their knowledge of evaluation techniques
and read about the social and psychological issues that underpin design of the web,
wireless, and collaborative systems are advised to read Chapter 1 for an overview,
then select from Chapters 10-14 on usability testing. Chapters 3,4, and 5 provide discussion of seminal user issues (cognitive, social, and affective aspects). There is new
material throughout the rest of the book, which will also be of interest for dipping
into as needed. This group may also be particularly interested in Chapter 8 which, together with material on the book website, provides practical design examples.
Acknowledgements
Many people have helped to make this book a reality. We have benefited from the
advice and support of our many professional colleagues across the world, our students, friends, and families and we thank you all. We also warmly thank the following
people for reviewing the manuscript and making many helpful suggestions for improvements: Liam Bannon, Sara Bly, Penny Collings, Paul Dourish, Jean Gasen,
Peter Gregor, Stella Mills, Rory O'Connor, Scott Toolson, Terry Winograd, Richard
Furuta, Robert J.K. Jacob, Blair Nonnecke, William Buxton, Carol Traynor, Blaise
Liffich, Jan Scott, Sten Hendrickson, Ping Zhang, Lyndsay Marshall, Gary Perlman,
Andrew Dillon, Michael Harrison, Mark Crenshaw, Laurie Dingers, David Carr,
Steve Howard, David Squires, George Weir, Marilyn Tremaine, Bob Fields, Frances
Slack, Ian Graham, Alan O'Callaghan, Sylvia Wilbur, and several anonymous reviewers. We also thank Geraldine Fitzpatrick, Tim and Dirk from DSTC (Australia)
for their feedback on Chapters 1 and 4, Mike Scaife, Harry Brignull, Matt Davies,
the HCCS Masters students at Sussex University (2000-2001), Stephanie Wilson
and the students from the School of Informatics at City University and Information
Systems Department at UMBC for their comments.
We are particularly grateful to Sara Bly, Karen Holtzblatt, Jakob Nielsen, Abigail Sellen, Suzanne Robertson, Gitta Salomon, Ben Shneiderman, Gillian Crampton Smith, and Terry Winograd for generously contributing in-depth interviews.
Lili Cheng and her colleagues allowed us to use the Hutchworld case study.
Bill Killam provided the TRZS case study. Keith Cogdill supplied the MEDLZNEplus case study. We thank Lili, Bill, and Keith for supplying the basic reports and
commenting on various drafts. Jon Lazar and Dorine Andrews contributed material for the section on questionnaires, which we thank them for.
We are grateful to our Editors Paul Crockett and Gaynor Redvers-Mutton and
the production team at Wiley: Maddy Lesure, Susannah Barr, Anna Melhorn,
Gemma Quilter, and Ken Santor. Without their help and skill this book would not
have been produced. Bill Zobrist and Simon Plumtree played a significant role in
persuading us to work with Wiley and we thank them too.
About the authors
I
xi
About the authors
The authors are all senior academics with a background in teaching, researching,
and consulting in the UK, USA, Canada, Australia, and Europe. Having worked
together on two other successful text books, they bring considerable experience in
curriculum development, using a variety of media for distance learning as well as
face-to-face teaching. They have considerable knowledge of creating learning texts
and websites that motivate and support learning for a range of students.
All three authors are specialists in interaction design and human-computer interaction (HCI). In addition they bring skills from other discipline~.Yvonne
Rogers is a cognitive scientist, Helen Sharp is a software engineer, and Jenny
Preece works in information systems. Their complementary knowledge and skills
enable them to cover the breadth of concepts in interaction design and HCI to produce an interdisciplinary text and website. They have collaborated closely, supporting and commenting upon each other's work to produce a high degree of
integration of ideas with one voice. They have shared everything from initial concepts, through writing, design and production.
Contents
Chapter 1
What is interaction design?
1.I
Introduction 1
1.2
1.3
1.4
1.5
1.6
Chapter 2
Good and poor design 2
1.2.1
What to design 4
What is interaction design? 6
1.3.1
The makeup of interaction design 6
1.3.2
Working together as a multidisciplinary team 9
1.3.3 Interaction design in business 10
What is involved in the process of interaction design? 12
The goals of interaction design 13
1.5.1
Usability goals 1A
1.5.2
User experience goals 18
More on usability: design and usability principles 20
1.6.1
Heuristics and usability principles 26
Interview with Gitta Salomon 3 1
Understanding and concep~alizing
interaction 35
2.1
2.2
2.3
2.4
2.5
2.6
Chapter 3
1
lntroduction 35
Understanding the problem space 36
Conceptual models 39
2.3.1
Conceptual models based on activities 41
2.3.2
Conceptual models based on objects 51
2.3.3
A case of mix and match? 54
Interface metaphors 55
Interaction paradigms 60
From conceptual models to physical design 64
Interview with Terry Winograd 70
Understanding users 73
3.1
3.2
3.3
3.4
Introduction 73
What is cognition? 74
Applying knowledge from the physical world to the digital world 90
Conceptual frameworks for cognition 92
3.4.1
Mental models 92
xiv Contents
3.5
3.4.2
Information processing 96
3.4.3
External cognition 98
Informing design: from theory to practice
101
Chapter 4
Designing for collaboration and communica~ion 105
4.1
Introduction 105
4.2
Social mechanisms used in communication and collaboration 106
4.2.1
Conversational mechanisms 107
Designing collaborative technologies to support conversation
4.2.2
110
4.2.3
Coordination mechanisms 118
Designing collaborative technologies to support coordination
4.2.4
122
4.2.5
Awareness mechanisms 124
4.2.6
Designing collaborative technologies to support awareness 126
4.3
Ethnographic studies of collaboration and communication 129
4.4
Conceptual frameworks 130
4.4.1
The language/action framework 130
4.4.2
Distributed cognition 133
Interview with Abigail Sellen 138
Chapter 5
Understanding how interfaces affect users 141
5.1
lntroduction 141
5.2
What are affective aspects? 142
5.3
Expressive interfaces 143
5.4
User frustration 147
5.4.1
Dealing with user frustration 152
A debate: the application of anthropomorphism to interaction design
5.5
5.6
Virtual characters: agents 157
5.6.1
Kinds of agents 157
5.6.2
General design concerns 160
Chapter 6
The process of interaction design
6.1
6.2
6.3
153
165
Introduction 165
What is interaction design about? 166
6.2.1
Four basic activities of interaction design 1 68
6.2.2
Three key characteristics of the interaction design process
Some practical issues 170
6.3.1
Who are the users? 171
170
Contents
6.4
Chapter
1
7
Chapter 8
6.3.2
What do we mean by "needs"? 172
6.3.3
How do you generate alternative designs? 174
6.3.4
How do you choose among alternative designs? 179
Lifecycle models: showing how the activities are related I 82
6.4.1
A simple lifecycle model for interaction design 186
6.4.2
Lifecycle models in software engineering 187
6.4.3
Lifecycle models in HCI 192
Interview with Gillian Crampton Smith 198
Identifying needs and establishing requirements 201
7.1
Introduction 201
7.2
What, how, and why? 202
7.2.1
What are we trying to achieve in this design activity? 202
7.2.2
How can we achieve this? 202
7.2.3
Why bother? The importance of getting it right 203
7.2.4
Why establish requirements? 204
7.3
What are requirements? 204
7.3.1
Different kinds of requirements 205
7.4
Data gathering 210
7.4.1
Data-gathering techniques 21 1
7.4.2
Choosing between techniques 215
7.4.3
Some basic datmgathering guidelines 216
7.5
Data interpretation and analysis 219
7.6
Task description 222
7.6.1
Scenarios 223
7.6.2
Use cases 226
7.6.3
Essential use cases 229
7.7
Task analysis 231
7.7.1
Hierarchical Task Analysis (HTA) 231
Interview with Suzanne Robertson 236
Design, prototyping and construction 239
8.1
lntroduction 239
8.2
Prototyping and construction 240
8.2.1
What is a prototype? 240
8.2.2
Why prototype? 241
8.2.3
Low-fidelity prototyping 243
8.2.4
High-fidelity prototyping 245
8.2.5
Compromises in prototyping 246
xv
xvi
Contents
8.3
8.4
8.5
Chapter 9
User-centered approaches to interaction design 279
9.1
9.2
9.3
9.4
9.5
Chapter 1 0
Introduction 279
Why is it important to involve users at all? 280
9.2.1
Degrees of involvement 281
What i s a user-centered approach? 285
Understanding users' work: applying ethnography in design
9.4.1
Coherence 293
9.4.2
Contextual Design 295
involving users in design: Participatory Design 306
9.5.1
PICTIVE 307
9.5.2
CARD 309
Interview with Karen Holtzblatt 313
288
Introducing evaluation 317
10.1 Introduction 317
10.2
10.3
10.4
Chapter 1 1
8.2.6
Construction: from design to implementation 248
Conceptual design: moving from requirements to first design 249
8.3.1
Three perspectives for developing a conceptual model 250
8.3.2
Expanding the conceptual model 257
8.3.3
Using scenarios in conceptual design 259
8.3.4
Using prototypes in conceptual design 262
Physical design: getting concrete 264
8.4.1
Guidelines for physical design 266
8.4.2
Different kinds of widget 268
Tool support 275
What, why, and when to evaluate 318
10.2.1 What to evaluate 318
10.2.2 Why you need to evaluate 319
10.2.3 When to evaluate 323
Hutchworld case study 324
10.3.1 How the team got started: early design ideas
10.3.2 How was the testing done? 327
10.3.3 Was it tested again? 333
10.3.4 Looking to the future 334
Discussion 336
An evaluation framework 339
1 1 .1
Introduction
339
324
Contents xvii
11.2
11.3
11.4
Evaluation paradigms and techniques 340
11.2.1 Evaluation paradigms 341
11.2.2 Techniques 345
D E C I D E: A framework to guide evaluation 348
11.3.1 Determine the goals 348
11.3.2 Explore the questions 349
11.3.3 Choose the evaluation paradigm and techniques
11.3.4 identify the practical issues 350
1 1.3.5 Decide how to deal with the ethical issues 351
11.3.6 Evaluate, interpret, and present the data 355
pilot studies 356
Chapter 12
Observing users
Chapter 13
Asking users and experts 389
349
359
12.1 Introduction 359
12.2 Goals, questions and paradigms 360
12.2.1 What and when to observe 361
12.2.2 Approaches to observation 363
1 2.3 How to observe 364
12.3.1 In controlled environments 365
12.3.2 In the field 368
12.3.3 Participant observation and ethnography 370
12.4 Data collection 373
12.4.1 Notes plus still camera 374
12.4.2 Audio recording plus still camera 374
12.4.3 Video 374
12.5 Indirect observation: tracking users' activities 377
12.5.1 Diaries 377
12.5.2 Interaction logging 377
12.6 Analyzing, interpreting and presenting data 379
12.6.1 Qualitative analysis to tell a story 380
1 2.6.2 Qualitative analysis for categorization 381
12.6.3 Quantitative data analysis 384
12.6.4 Feeding the findings back into design 384
Interview with Sara Bb 387
13.1
13.2
introduction 389
Aking users: interviews 390
13.2.1 Developing questions and planning an interview 390
xviii
Contents
13.3
13.4
13.5
Chapter 14
13.2.2 Unstructured interviews 392
13.2.3 Structured interviews 394
13.2.4 Semi-structured interviews 394
13.2.5 Group interviews 396
13.2.6 Other sources of interview-like feedback
13.2.7 Data analysis and interpretation 398
Asking users: Questionnaires 398
13.3.1 Designing questionnaires 398
13.3.2 Question and response format 400
13.3.3 Administering questionnaires 404
13.3.4 Online questionnaires 405
13.3.5 Analyzing questionnaire data 407
Asking experts: Inspections 407
13.4.1 Heuristic evaluation 408
13.4.2 Doing heuristic evaluation 410
13.4.3 Heuristic evaluation of websites 412
13.4.4 Heuristics for other devices 419
Asking experts: walkthroughs 420
I 3.5.1 Cognitive walkthroughs 420
13.5.2 Pluralistic walkthroughs 423
Interview with Jakob Nielsen 426
397
Testing and modeling users 429
14.1
14.2
14.3
14.4
Introduction 429
User testing 430
14.2.1 Testing MEDLINE~~us432
Doing user testing 438
14.3.1 Determine the goals and explore the questions 439
14.3.2 Choose the paradigm and techniques 439
14.3.3 Identify the practical issues: Design typical tasks 439
14.3.4 Identify the practical issues: Select typical users 440
14.3.5 Identify the practical issues: Prepare the testing
conditions 441
14.3.6 Identify the practical issues: Plan how to run the tests 442
14.3.7 Deal with ethical issues 443
14.3.8 Evaluate, analyze, and present the data 443
Experiments 443
14.4.1 Variables and conditions 444
14.4.2 Allocation of participants to conditions 445
Contents
1 4.5
Chapter 15
14.4.3 Other
issues 446
14.4.4 Data collection and analysis 446
Predictive models 448
14.5.1 The W M S model 449
14.5.2 The Keystroke level model 450
14.5.3 Benefits and limitations of W M S 453
14.5.4 Fitts' Law 454
Interview with Ben Shneiderman 457
Design and evaluation in the real world: communicators
and advisory systems 461
15.1
15.2
15.3
15.4
Introduction 461
Key Issues 462
Designing mobile communicators 463
15.3.1 Background 463
15.3.2 Nokia's approach to developing a communicator 464
15.3.3 Philip's approach to designing a communicator for children
474
Redesigning part of a large interactive phone-based response system 482
15.4.1 Background 483
15.4.2 The redesign 483
Reflections from the Authors 491
References 493
Credits
503
Index 509
xix
I
by Gary Perlman
As predicted by many visionaries, devices everywhere are getting "smarter." My
camera has a multi-modal hierarchical menu and form interface. Even my toaster
has a microprocessor. Computing is not just for computers anymore. So when the
authors wrote the subtitle "beyond human-computer interaction," they wanted to
convey that the book generalizes the human side to people, both individuals and
groups, and the computer side to desktop computers, handheld computers, phones,
cameras . . . maybe even toasters.
My own interest in this book is motivated by having been a software developer
for 20 years, during which time I was a professor and consultant for 12. Would the
book serve as a textbook for students? Would it help bring software development
practice into a new age of human-centered interaction design?
A textbook for students . . .
More than anything, I think students need to be motivated, inspired, challenged,
and I think this book, particularly Chapters 1-5, will do that. Many students will
not have the motivating experience of seeing projects and products fail because of
a lack of attention, understanding, and zeal for the user, but as I read the opening
chapters, I imagined students thinking, "This is what I've been looking for!" The interviews will provide students with the wisdom of well-chosen experts: what's important, what worked (or didn't), and why. I see students making career choices
based on this motivating material.
The rest of the book covers the art and some of the science of interaction design, the basic knowledge needed by practitioners and future innovators. Chapters
6-9 give a current view of analysis, design, and prototyping, and the book's website
should add motivating examples. Chapters 10-14 cover evaluation in enough depth
to facilitate understanding, not just rote application. Chapter 15 brings it all together, adding more depth. For each topic, there are ample pointers to further
reading, which is important because interaction design is not a one-book discipline.
Finally, the book itself is pedagogically well designed. Each chapter describes
its aims, contains examples and subtopics, and ends with key points, assignments,
and an annotated bibliography for more detail.
A guide for development teams . . .
When I lead or consult on software projects, I face the same problem over and over:
many people in marketing and software development-these are the people who
have the most input into design, but it applies to any members of multidisciplinary
teams-have little knowledge or experience building systems with a user-centered
xxii
Foreword
focus. A user-centered focus requires close work with users (not just customer-buyers), from analysis through design, evaluation, and maintenance. A lack of usercentered focus results in products and services that often do not meet the needs of
their intended users. Don Norman's design books have convinced many that these
problems are not unique to software, so this book's focus on interaction design feels
right.
To help software teams adopt a user-centered focus, I've searched for books
with end-to-end coverage from analysis, to design, to implementation (possibly of
prototypes), to evaluation (with iteration). Some books have tried to please all audiences and have become encyclopedias of user interface development, covering
topics worth knowing, but not in enough detail for readers to understand them.
Some books have tried to cover theory in depth and tried to appeal to developers
who have little interest in theory. Whatever the reasons for these choices, the results have been lacking. This book has chosen fewer topics and covered them in
more depth; enough depth, I think, to put the ideas into practice. I think the material is presented in a way that is understandable by a wide audience, which is important in order for the book to be useful to whole multidisciplinary teams.
A recommended book . . .
I've been waiting for this book for many years. I think it's been worth the wait.
As the director of the HCI Bibliography project (www.hcibib.org), a free-access HCI portal receiving a half-million hits per year, I receive many requests for
suggestions for books, particularly from students and software development managers. To answer that question, I maintain a list of recommended readings in ten
categories (with 20,000 hits per year). Until now, it's been hard to recommend just
one book from that list. I point people to some books for motivation, other books
for process, and books for specific topics (e.g., task analysis, ergonomics, usability
testing). This book fits well into half the categories in my list and makes it easier to
recommend one book to get started and to have on hand for development.
I welcome the commitment of the authors to building a website for the book.
It's a practice that has been adopted by other books in the field to offer additional
information and keep the book current. The site also presents interactive content
to aid in tasks like conducting surveys and heuristic evaluations. I look forward to
seeing the book's site present new materials, but as director of www.hcibib.org, I
hope they use links to instead of re-inventing existing resources.
Gary Perlman
Columbus
October 2001
Foreword
xxiii
About Gary Perlman
Gary Perlman is a consulting research scientist at the OCLC-Online Computer Library Center (www.oclc.org) where he works on user interfaces for bibliographic
and full-text retrieval. His research interests are in making information technology
more useful and usable for people.
He has also held research and academic positions at Bell Labs in Murray Hill,
New Jersey; Wang Institute of Graduate Studies; Massachusetts Institute of Technology; Carnegie-Mellon University; and The Ohio State University. Dr. Perlman's
Ph.D. is in experimental psychology from the University of California, San Diego.
He is the author of over 75 publications in the areas of mathematics education, statistical computing, hypertext, and user interface development. He has lectured and
consulted internationally since 1980.
He is best known in the HCI community as the director of the HCI Bibliography (www.hcibib.org), a free-access online resource of over 20,000 records
searched hundreds of thousands of times each year.
A native of Montreal, Canada, Gary now lives in Columbus, Ohio with his wife
and two sons.
What is interaction design?
1 .I Introduction
1.2 Good and poor design
1.2.1 What to design
1.3 What is interaction design?
1.3.1 The makeup of interaction design
1.3.2 Working together as a multidisciplinary team
1 3.3 Interaction design in business
1.4 What is involved in the process of interaction design?
1.5 The goals of interaction design
1.5.1Usability goals
1.5.2User experience goals
1.6.More on usability: design and usability principles
1.1
Introduction
How many interactive products are there in everyday use? Think for a minute
about what you use in a typical day: cell phone, computer, personal organizer, remote control, soft drink machine, coffee machine, ATM, ticket machine, library information system, the web, photocopier, watch, printer, stereo, calculator, video
game.. . the list is endless. Now think for a minute about how usable they are.
How many are actually easy, effortless, and enjoyable to use? All of them, several,
or just one or two? This list is probably considerably shorter. Why is this so?
Think about when some device caused you considerable grief-how much time
did you waste trying to get it to work? Two well-known interactive devices that
cause numerous people immense grief are the photocopier that doesn't copy the
way they want and the VCR that records a different program from the one they
thought they had set or none at all. Why do you think these things happen time and
time again? Moreover, can anything be done about it?
Many products that require users to interact with them to carry out their tasks
(e.g., buying a ticket online from the web, photocopying an article, pre-recording a TV
program) have not necessarily been designed with the users in mind. Typically, they
have been engineered as systems to perform set functions. While they may work effectively from an engineering perspective, it is often at the expense of how the system will
be used by real people. The aim of interaction design is to redress this concern by
2
Chapter 1
What is interaction design?
bringing usability into the design process. In essence, it is about developing interactive
products1 that are easy, effective, and enjoyable to use-from the users' perspective.
In this chapter we begin by examining what interaction design is. We look at
the difference between good and poor design, highlighting how products can differ
radically in their usability. We then describe what and who is involved in interaction design. In the last part of the chapter we outline core aspects of usability and
how these are used to assess interactive products. An assignment is presented at
the end of the chapter in which you have the opportunity to put into practice what
you have read, by evaluating an interactive product using various usability criteria.
The main aims of the chapter are to:
Explain the difference between good and poor interaction design.
Describe what interaction design is and how it relates to human-computer
interaction and other fields.
Explain what usability is.
Describe what is involved in the process of interaction design.
Outline the different forms of guidance used in interaction design.
Enable you to evaluate an interactive product and explain what is good and
bad about it in terms of the goals and principles of interaction design.
1.2 Good and poor design
A central concern of interaction design is to develop interactive products that are
usable. By this is generally meant easy to learn, effective to use, and provide an enjoyable user experience. A good place to start thinking about how to design usable
interactive products is to compare examples of well and poorly designed ones.
Through identifying the specific weaknesses and strengths of different interactive
systems, we can begin to understand what it means for something to be usable or
not. Here, we begin with an example of a poorly designed system-voice mailthat is used in many organizations (businesses, hotels, and universities). We then
compare this with an answering machine that exemplifies good design.
Imagine the following scenario. You're staying at a hotel for a week while on a
business trip. You discover you have left your cell (mobile) phone at home so you
have to rely on the hotel's facilities. The hotel has a voice-mail system for each
room. To find out if you have a message, you pick up the handset and listen to the
tone. If it goes "beep beep beep" there is a message. To find out how to access the
message you have to read a set of instructions next to the phone.
You read and follow the first step:
"1. Touch 491".
The system responds, "You have reached the Sunny Hotel voice message center.
Please enter the room number for which you would like to leave a message."
'We use the term interactive products generically to refer to all classes of interactive systems,
technologies, environments, tools, applications,and devices.
1.2 Good and poor design
3
You wait to hear how to listen to a recorded message. But there are no further
instructions from the phone. You look down at the instruction sheet again and
read:
"2. Touch*, your room number, and #". You do so and the system replies,
"You have reached the mailbox for room 106. To leave a message type in your
password."
You type in the room number again and the system replies, "Please enter room
number again and then your password."
You don't know what your password is. You thought it was the same as your
room number. But clearly not. At this point you give up and call reception for help.
The person at the desk explains the correct procedure for recording and listening
to messages. This involves typing in, at the appropriate times, the room number
and the extension number of the phone (the latter is your password, which is different from the room number). Moreover, it takes six steps to access a message and
five steps to leave a message. You go out and buy a new cell phone.
What is problematic with this voice-mail system?
It is infuriating.
It is confusing.
It is inefficient, requiring you to carry out a number of steps for basic tasks.
It is difficult to use.
It has no means of letting you know at a glance whether any messages have
been left or how many there are. You have to pick up the handset to find out
and then go through a series of steps to listen to them.
It is not obvious what to do: the instructions are provided partially by the
system and partially by a card beside the phone.
Now consider the following phone answering machine. Figure 1.1 shows two
small sketches of an answering machine phone. Incoming messages are represented
using physical marbles. The number of marbles that have moved into the pinballlike chute indicates the number of messages. Dropping one of these marbles into a
slot in the machine causes the recorded message to play. Dropping the same marble into another slot on the phone dials the caller who left the message.
Figure 1 .1 Two small
sketches showing answering phone.
4
Chapter 1
What is interaction design?
How does the "marble" answering machine differ from the voice-mail system?
It uses familiar physical objects that indicate visually at a glance how many
messages have been left.
It is aesthetically pleasing and enjoyable to use.
It only requires one-step actions to perform core tasks.
It is a simple but elegant design.
It offers less functionality and allows anyone to listen to any of the messages.
The marble answering machine was designed by Durrell Bishop while a student at the Royal College of Art in London (described by Crampton-Smith, 1995).
One of his goals was to design a messaging system that represented its basic functionality in terms of the behavior of everyday objects. To do this, he capitalized on
people's everyday knowledge of how the physical world works. In particular, he
made use of the ubiquitous everyday action of picking up a physical object and
putting it down in another place. This is an example of an interactive product designed with the users in mind. The focus is on providing them with an enjoyable experience but one that also makes efficient the activity of receiving messages.
However, it is important to note that although the marble answering machine is a
very elegant and usable design, it would not be practical in a hotel setting. One of
the main reasons is that it is not robust enough to be used in public places, for example, the marbles could easily get lost or taken as souvenirs. Also, the need to
identify the user before allowing the messages to be played is essential in a hotel
setting. When considering the usability of a design, therefore, it is important to
take into account where it is going to be used and who is going to use it. The marble
answering machine would be more suited in a home setting-provided there were
no children who might be tempted to play with the marbles!
1.2.1 What to design
Designing usable interactive products thus requires considering who is going to be
using them and where they are going to be used. Another key concern is understanding the kind of activities people are doing when interacting with the products.
The appropriateness of different kinds of interfaces and arrangements of input and
output devices depends on what kinds of activities need to be supported. For example, if the activity to be supported is to let people communicate with each other at a
distance, then a system that allows easy input of messages (spoken or written) that
can be readily accessed by the intended recipient is most appropriate. In addition,
an interface that allows the users to interact with the messages (e.g., edit, annotate,
store) would be very useful.
The range of activities that can be supported is diverse. Just think for a
minute what you can currently do using computer-based systems: send messages,
gather information, write essays, control power plants, program, draw, plan, calculate, play games-to name but a few. Now think about the number of interfaces and interactive devices that are available. They, too, are equally diverse:
1.2 Good and poor design
5
multimedia applications, virtual-reality environments, speech-based systems, personal digital assistants and large displays-to name but a few. There are also
many ways of designing the way users can interact with a system (e.g., via the use
of menus, commands, forms, icons, etc.). Furthermore, more and more novel
forms of interaction are appearing that comprise physical devices with embedded
computational power, such as electronic ink, interactive toys, smart fridges, and
networked clothing (See Figure 1.2 on Color Plate 1). What this all amounts to is
a multitude of choices and decisions that confront designers when developing interactive products.
A key question for interaction design is: how do you optimize the users' interactions with a system, environment or product, so that they match the users' activities that are being supported and extended? One could use intuition and hope for
the best. Alternatively, one can be more principled in deciding which choices to
make by basing them on an understanding of the users. This involves:
taking into account what people are good and bad at
considering what might help people with the way they currently do things
thinking through what might provide quality user experiences
listening to what people want and getting them involved in the design
using "tried and tested" user-based techniques during the design process
The aim of this book is to cover these aspects with the goal of teaching you how to
carry out interaction design. In particular, it focuses on how to identify users'
needs, and from this understanding, move to designing usable, useful, and enjoyable systems.
How does making a phone call differ when using:
a public phone box
a cell phone?
How have these devices been designed to take into account (a) the kind of users, (b) type
of activity being supported, and (c) context of use?
Comment
(a) Public phones are designed to be used by the general public. Many have Braille embossed on the keys and speaker volume control to enable people who are blind and
hard of hearing to use them.
Cell phones are intended for all user groups, although they can be difficult to use for
people who are blind or have limited manual dexterity.
(b) Most phone boxes are designed with a simple mode of interaction: insert card or
money and key in the phone number. If engaged or unable to connect the money or
card is returned when the receiver is replaced. There is also the option of allowing the
caller to make a follow-on call by pressing a button rather than collecting the money
and reinserting it again. This function enables the making of multiple calls to be more
efficient.
I
6 Chapter 1
What is interaction design?
Cell phones have a more complex mode of interaction. More functionality is provided,
requiring the user to spend time learning how to use them. For example, users can save
phone numbers in an address book and then assign these to "hotkeys," allowing them
to be called simply through pressing one or two keys.
(c) Phone boxes are intended to be used in public places, say on the street or in a bus station, and so have been designed to give the user a degree of privacy and noise protection through the use of hoods and booths.
Cell phones have have been designed to be used any place and any time. However, little consideration has been given to how such flexibility affects others who may be in
the same public place (e.g.,sitting on trains and buses).
I
I
I
1.3 What is interaction design?
By interaction design, we mean
designing interactive products to support people in their everyday and working lives.
In particular, it is about creating user experiences that enhance and extend the way
people work, communicate and interact. Winograd (1997) describes it as "the design of spaces for human communication and interaction." In this sense, it is about
finding ways of supporting people. This contrasts with software engineering, which
focuses primarily on the production of software solutions for given applications. A
simple analogy to another profession, concerned with creating buildings, may clarify this distinction. In his account of interaction design, Terry Winograd asks how
architects and civil engineers differ when faced with the problem of building a
house. Architects are concerned with the people and their interactions with each
other and within the house being built. For example, is there the right mix of family
and private spaces? Are the spaces for cooking and eating in close proximity? Will
people live in the space being designed in the way it was intended to be used? In
contrast, engineers are interested in issues to do with realizing the project. These
include practical concerns like cost, durability, structural aspects, environmental
aspects, fire regulations, and construction methods. Just as there is a difference
between designing and building a house, so too, is there a distinction between interaction design and software engineering. In a nutshell, interaction design is related to software engineering in the same way as architecture is related to civil
engineering.
1.3.1 The makeup of interaction design
It has always been acknowledged that for interaction design to succeed many disciplines need to be involved. The importance of understanding how users act and
react to events and how they communicate and interact together has led people
from a variety of disciplines, such as psychologists and sociologists, to become involved. Likewise, the growing importance of understanding how to design different
kinds of interactive media in effective and aesthetically pleasing ways has led to a
1.3 What is interaction design?
7
diversity of other practitioners becoming involved, including graphic designers,
artists, animators, photographers, film experts, and product designers. Below we
outline a brief history of interaction design.
In the early days, engineers designed hardware systems for engineers to use.
The computer interface was relatively straightforward, comprising various switch
panels and dials that controlled a set of internal registers. With the advent of monitors (then referred to as visual display units or VDUs) and personal workstations in
the late '70s and early '80s, interface design came into being (Grudin, 1990). The
new concept of the user interface presented many challenges:
Terror. You have to confront the documentation. You have to learn a new language. Did
you ever use the word 'interface' before you started using the computer?
-Advertising executive Arthur Einstein (1990)
One of the biggest challenges at that time was to develop computers that could
be accessible and usable by other people, besides engineers, to support tasks involving human cognition (e.g., doing sums, writing documents, managing accounts,
drawing plans). To make this possible, computer scientists and psychologists became involved in designing user interfaces. Computer scientists and software engineers developed high-level programming languages (e.g., BASIC, Prolog), system
architectures, software design methods, and command-based languages to help in
such tasks, while psychologists provided information about human capabilities
(e.g., memory, decision making).
The scope afforded by the interactive computing technology of that time (i.e.,
the combined use of visual displays and interactive keyboards) brought about
many new challenges. Research into and development of graphical user interfaces (GUI for short, pronounced "goo-ee") for office-based systems took off in
a big way. There was much research into the design of widgets (e.g., menus, windows, palettes, icons) in terms of how best to structure and present them in a
GUI.
In the mid '80s, the next wave of computing technologies-including speech
recognition, multimedia, information visualization, and virtual reality-presented
even more opportunities for designing applications to support even more people.
Education and training were two areas that received much attention. Interactive
learning environments, educational software, and training simulators were some of
the main outcomes. To build these new kinds of interactive systems, however, required a different kind of expertise from that of psychologists and computer programmers. Educational technologists, developmental psychologists, and training
experts joined in the enterprise.
As further waves of technological development surfaced in the '90s-networking, mobile computing, and infrared sensing-the creation of a diversity of applications for all people became a real possibility. All aspects of a person's life-at
home, on the move, at school, at leisure as well as at work, alone, with family or
friends-began to be seen as areas that could be enhanced and extended by designing and integrating various arrangements of computer technologies. New ways of
learning, communicating, working, discovering, and living were envisioned.
8
Chapter 1
What is interaction design?
In the mid '90s, many companies realized it was necessary again to extend their
existing multidisciplinary design teams to include professionals trained in media
and design, including graphical design, industrial design, film, and narrative. Sociologists, anthropologists, and dramaturgists were also brought on board, all having
quite a different take on human interaction from psychologists. This wider set of
1.3 What is interaction design?
9
people were thought to have the right mix of skills and understanding of the different application areas necessary to design the new generation of interactive systems.
For example, designing a reminder application for the family requires understanding how families interact; creating an interactive story kit for children requires understanding how children write and understand narrative, and developing an
interactive guide for art-gallery visitors requires appreciating what people do and
how they move through public spaces.
Now in the 'OOs, the possibilities afforded by emerging hardware capabilitiese.g., radio-frequency tags, large interactive screens, and information applianceshas led to a further realization that engineers, who know about hardware, software,
and electronics are needed to configure, assemble, and program the consumer electronics and other devices to be able to communicate with each other (often referred to as middleware).
1.3.2 Working together as a multidisciplinary team
Bringing together so many people with different backgrounds and training has
meant many more ideas being generated, new methods being developed, and more
creative and original designs being produced. However, the down side is the costs
involved. The more people there are with different backgrounds in a design team,
the more difficult it can be to communicate and progress forward the designs being
generated. Why? People with different backgrounds have different perspectives
and ways of seeing and talking about the world (see Figure 1.4). What one person
values as important others may not even see (Kim, 1990). Similarly, a computer scientist's understanding of the term representation is often very different from a
graphic designer's or a psychologist's.
Four different
team members looking at
the same square, but each
seeing it quite differently.
Figure 1.4
10
Chapter 1
What is interaction design?
What this means in practice is that confusion, misunderstanding, and communication breakdowns can often surface in a team. The various team members
may have different ways of talking about design and may use the same terms to
mean quite different things. Other problems can arise when a group of people is
"thrown" together who have not worked as a team. For example, the Philips Vision of the Future Project found that its multidisciplinary teams-who were responsible for developing ideas and products for the future-experienced a
number of difficulties, namely, that project team members did not always have a
clear idea of who needed what information, when, and in what form (Lambourne
et al., 1997).
practice, the makeup of a given design team depends on the kind of interactive product
ing built. Who do you think would need to be involved in developing:
(a) a public kiosk providing information about the exhibits available in a science
museum?
(b) an interactive educational website to accompany a TV series?
Comment
Each team will need a pumber of different people with different skill sets. For example, the
first interactive product would need:
(a) graphic and inteiaction designers, museum curators, educational advisors, software
engineers, software designers, usability engineers, ergonomists
The second project would need:
(b) TV producers, graphic and interaction designers, teachers, video experts, software
engineers, software designers, usability engineers
In addition, as both systeds are being developed for use by the general public, representative users, such as school children and parents, should be involved.
In practice, design teams often end up being quite large, especially if they are working on a
big project to meet a fixed deadline. For example, it is common to find teams of fifteen people or more working on a website project for an extensive period of time, like six months.
This means that a number of people from each area of expertise are likely to be working as
part of the project team.
1.3.3 Interaction design in business
Interaction design is dbw big business. In particular, website consultants, startup companies, a n d mobile computing industries have all realized its pivotal role
in successful interactive hroducts. To get noticed in the highly competitive field
of web products requires standing out. Being able to say that your product is
easy and effective to use is seen as central to this. Marketing departments are realizing how branding, the number of hits, customer return rate, and customer
satisfaction are greatly affected by the usability of a website. Furthermore, the
presence or absence of good interaction design can make or break a company.
1.3 What is interaction design?
11
One infamous dot.com fashion clothes company that failed to appreciate the importance of good interaction design paid heavily for its oversight, becoming
bankrupt within a few months of going public.' Their approach had been to go
for an "all singing and all dancing," glossy 3D graphical interface. One of the
problems with this was that it required several minutes to download. Furthermore, it often took more than 20 minutes to place an order by going through a
painfully long and slow process of filling out an online form-only to discover
that the order was not successful. Customers simply got frustrated with the site
and never returned.
In response to the growing demand for interaction design, an increasing
number of consultancies are establishing themselves as interaction design experts. One such company is Swim, set up by Gitta Salomon to assist clients with
the design of interactive products (see the interview with her at the end of this
chapter). She points out how often companies realize the importance of interaction design but don't know how to do it themselves. So they get in touch with
companies, like Swim, with their partially developed products and ask them for
help. This can come in the form of an expert "crit" in which a detailed review of
the usability and design of the product is given (for more on expert evaluation,
see Chapter 13). More extensively, it can involve helping clients create their
products.
Another established design company that practices interaction design is IDEO,
which now has many branches worldwide. Drawing on over 20 years of experience
in the area, they design products, services, and environments for other companies,
pioneering new user experiences (Spreenberg et al., 1995). They have developed
'This happened before the dot.com crash in 2001.
12
Chapter 1
What is interaction design?
Figure 1.5 An innovative
product developed by
IDEO: Scout Modo, a wire-
less handheld device delivering up-to-date
information about what's
going on in a city.
thousands of products for numerous clients, each time following their particular
brand of user-centered design (see Figure 1.5).
1.4
What is involved in the process of interaction design?
Essentially, the process of interaction design involves four basic activities:
1. Identifying needs and establishing requirements.
2. Developing alternative designs that meet those requirements.
3. Building interactive versions of the designs so that they can be communicated and assessed.
4. Evaluating what is being built throughout the process.
These activities are intended to inform one another and to be repeated. For example, measuring the usability of what has been built in terms of whether it is easy to
use provides feedback that certain changes must be made or that certain requirements have not yet been met.
Evaluating what has been built is very much at the heart of interaction design.
Its focus is on ensuring that the product is usable. It is usually addressed through a
user-centered approach to design, which, as the name suggests, seeks to involve
users throughout the design process. There are many different ways of achieving
this: for example, through observing users, talking to them, interviewing them, testing them using performance tasks, modeling their performance, asking them to fill
1.5 The goals of interaction design
13
in questionnaires, and even asking them to become co-designers. The findings from
the different ways of engaging and eliciting knowledge from users are then interpreted with respect to ongoing design activities (we give more detail about all these
aspects of evaluation in Chapters 10-14).
Equally important as involving users in evaluating an interactive product is understanding what people currently do. This form of research should take place before building any interactive product. Chapters 3,4, and 5 cover a lot of this ground
by explaining in detail how people act and interact with one another, with information, and with various technologies, together with describing their strengths and
weaknesses. Such knowledge can greatly help designers determine which solutions
to choose from the many design alternatives available and how to develop and test
these further. Chapter 7 describes how an understanding of users' needs can be
translated to requirements, while Chapter 9 explains how to involve users effectively in the design process.
A main reason for having a better understanding of users is that different
users have different needs and interactive products need to be designed accordingly. For example, children have different expectations about how they want
to learn or play from adults. They may find having interactive quizzes and cartoon
characters helping them along to be highly motivating, whereas most adults find
them annoying. Conversely, adults often like talking-heads discussions about topics, but children find them boring. Just as everyday objects like clothes, food, and
games are designed differently for children, teenagers, and adults, so, too, must interactive products be designed to match the needs of different kinds of users.
In addition to the four basic activities of design, there are three key characteristics of the interaction design process:
1. Users should be involved through the development of the project.
2. Specific usability and user experience goals should be identified, clearly documented, and agreed upon at the beginning of the project.
3. Iteration through the four activities is inevitable.
We have already mentioned the importance of involving users and will return to
this topic throughout the book. Iterative design will also be addressed later when
we talk about the various design and evaluation methods by which this can be
achieved. In the next section we describe usability and user experience goals.
1.5
The goals of interaction design
Part of the process of understanding users' needs, with respect to designing an interactive system to support them, is to be clear about your primary objective. Is it
to design a very efficient system that will allow users to be highly productive in
their work, or is it to design a system that will be challenging and motivating so that
it supports effective learning, or is it something else? We call these top-level concerns usability goals and user experience goals. The two differ in terms of how they
are operationalized, i.e., how they can be met and through what means. Usability
14
Chapter 1
What is interaction design?
goals are concerned with meeting specific usability criteria (e.g., efficiency) and
user experience goals are largely concerned with explicating the quality of the user
experience (e.g., to be aesthetically pleasing).
1.5.1 Usability goals
To recap, usability is generally regarded as ensuring that interactive products are
easy to learn, effective to use, and enjoyable from the user's perspective. It involves
optimizing the interactions people have with interactive products to enable them to
carry out their activities at work, school, and in their everyday life. More specifically, usability is broken down into the following goals:
effective to use (effectiveness)
efficient to use (efficiency)
safe to use (safety)
have good utility (utility)
easy to learn (learnability)
easy to remember how to use (memorability)
For each goal, we describe it in more detail and provide a key question.
Effectiveness is a very general goal and refers to how good a system is at doing
what it is supposed to do.
Question: Is the system capable of allowing people to learn well, carry out their
work efficiently, access the information they need, buy the goods they want, and
so on?
Efficiency refers to the way a system supports users in carrying out their tasks.
The answering machine described at the beginning of the chapter was considered
efficient in that it let the user carry out common tasks (e.g., listening to messages)
through a minimal number of steps. In contrast, the voice-mail system was considered inefficient because it required the user to carry out many steps and learn an
arbitrary set of sequences for the same common task. This implies that an efficient
way of supporting common tasks is to let the user use single button or key presses.
An example of where this kind of efficiency mechanism has been effectively employed is in e-tailing. Once users have entered all the necessary personal details on
an e-commerce site to make a purchase, they can let the site save all their personal
details. Then, if they want to make another purchase at that site, they don't have
to re-enter all their personal details again. A clever mechanism patented by
Amazon.com is the one-click option, which requires users only to click a single button when they want to make another purchase.
Question: Once users have learned how to use a system to carry out their tasks,
can they sustain a high level of productivity?
Safety involves protecting the user from dangerous conditions and undesirable
situations. In relation to the first ergonomic aspect, it refers to the external conditions where people work. For example, where there are hazardous conditions-like
X-ray machines or chemical plants--operators should be able to interact with and
control computer-based systems remotely. The second aspect refers to helping any
1.5 The goals of interaction design
15
kind of user in any kind of situation avoid the dangers of carrying out unwanted actions aceidentally. It also refers to the perceived fears users might have of the consequences of making errors and how this affects their behavior. To make
computer-based systems safer in this sense involves (i) preventing the user from
making serious errors by reducing the risk of wrong keyslbuttons being mistakenly
activated (an example is not placing the quit or delete-file command right next to
the save command on a menu) and (ii) providing users with various means of recovery should they make errors. Safe interactive systems should engender confidence and allow the user the opportunity to explore the interface to try out new
operations (see Figure 1.6a). Other safety mechanisms include undo facilities and
Color Settings
b
lb)
Figure 1.6 (a) A safe and an unsafe menu. Which is which and why? (b) Warning dialog
message from Eudora.
16
Chapter 1
What is interaction design?
confirmatory dialog boxes that give users another chance to consider their intentions (a well-known example used in e-mail applications is the appearance of a dialog box, after the user has highlighted messages to be deleted, saying: "Are you
sure you want to delete all these messages?" See Figure 1.6(b)).
Question: Does the system prevent users from making serious errors and, if
they do make an error, does it permit them to recover easily?
Utility refers to the extent to which the system provides the right kind of functionality so that users can do what they need or want to do. An example of a system
with high utility is an accounting software package providing a powerful computational tool that accountants can use to work out tax returns. A example of a system
with low utility is a software drawing tool that does not allow users to draw freehand but forces them to use a mouse to create their drawings, using only polygon
shapes.
Question: Does the system provide an appropriate set of functions that enable
users to carry out all their tasks in the way they want to do them?
Learnability refers to how easy a system is to learn to use. It is well known that
people don't like spending a long time learning how to use a system. They want to
get started straight away and become competent at carrying out tasks without too
much effort. This is especially so for interactive products intended for everyday use
(e.g., interactive TV, email) and those used only infrequently (e.g., videoconferencing). To a certain extent, people are prepared to spend longer learning more complex systems that provide a wider range of functionality (e.g., web authoring tools,
word processors). In these situations, CD-ROM and online tutorials can help by
providing interactive step-by-step material with hands-on exercises. However,
many people find these tedious and often difficult to relate to the tasks they want to
1.5 The goals of interaction design
17
accomplish. A key concern is determining how much time users are prepared to
spend learning a system. There seems little point in developing a range of functionality if the majority of users are unable or not prepared to spend time learning how
to use it.
Question: How easy is it and how long does it take (i) to get started using a system t o perform core tasks and (ii) to learn the range of operations to perform a
wider set of tasks?
Memorability refers to how easy a system is to remember how to use, once
learned. This is especially important for interactive systems that are used infrequently. If users haven't used a system or an operation for a few months or longer,
they should be able to remember or at least rapidly be reminded how to use it.
Users shouldn't have to keep relearning how to carry out tasks. Unfortunately, this
tends to happen when the operations required to be learned are obscure, illogical,
or poorly sequenced. Users need to be helped to remember how to do tasks. There
are many ways of designing the interaction to support this. For example, users can
be helped to remember the sequence of operations at different stages of a task
through meaningful icons, command names, and menu options. Also, structuring
options and icons so they are placed in relevant categories of options (e.g., placing
all the drawing tools in the same place on the screen) can help the user remember
where to look to find a particular tool at a given stage of a task.
Question: What kinds of interface support have been provided to help users remember how to carry out tasks, especially for systems and operations that are used
infrequently?
How long do you think it should take to learn how to use the following interactive products
and how long does it actually take most people to learn them? How memorable are they?
(a) using a VCR to play a video
(b) using a VCR to pre-record two programs
(c) using an authoring tool to create a website
Comment
(a) To play a video should be as simple as turning the radio on, should take less than 30
seconds to work out, and then should be straightforward to do subsequently. Most
people are able to fathom how to play a video. However, some systems require the
user to switch to the "video" channel using one or two remote control devices, selecting from a choice of 50 or more channels. Other settings may also need to be configured before the video will play. Most people are able to remember how to play a video
once they have used a particular VCR.
(b) This is a more complex operation and should take a couple of minutes to learn how to
do and to check that the programming is correct. In reality, many VCRs are so poorly
designed that 80% of the population is unable to accomplish this task, despite several
attempts. Very few people remember how to pre-record a program, largely because
the interaction required to do this is poorly designed, with poor or no feedback, and is
often illogical from the user's perspective. Of those, only a few will bother to go
through the manual again.
18
Chapter 1
Whpt is interaction design?
(c) A well-designed authoring too1 should let the user create a basic page in about 20 minutes. Learning the full range of operations and possibilities is likely to take much
longer, possibly a few days. In reality, there are some good authoring tools that allow
the user to get started straight away, providing templates that they can adapt. Most
users will extend their repertoire, taking another hour or so to learn more functions.
However, very few people actually learn to use the full range of functions provided by
the authoring tool. Users will tend to remember frequently used operations (e.g., cut
and paste, inserting images), especially if they are consistent with the way they are carried out in other software applications. However, less frequently used operations may
need to be relearned (e.g., formatting tables).
The usability goals discussed so far are well suited to the design of business systems
intended to support working practices. In particular, they are highly relevant for
companies and organizations who are introducing or updating applications running
on desktop and networked systems-that are intended to increase productivity by
improving and enhancing how work gets done. As well as couching them in terms
of specific questions, usability goals are turned into usability criteria. These are
specific objectives that enable the usability of a product to be assessed in terms of
how it can improve (or not) a user's performance. Examples of commonly used usability criteria are time to complete a task (efficiency), time to learn a task (learnability), and the number of errors made when carrying out a given task over time
(memorability).
1.5.2 User experience goals
The realization that new technologies are offering increasing opportunities for supporting people in their everyday lives has led researchers and practitioners to consider further goals. The emergence of technologies (e.g., virtual reality, the web,
mobile computing) in a diversity of application areas (e.g., entertainment, education, home, public areas) has brought about a much wider set of concerns. As well
as focusing primarily on improving efficiency and productivity at work, interaction
design is increasingly concerning itself with creating systems that are:
satisfying
enjoyable
fun
entertaining
helpful
motivating
aesthetically pleasing
supportive of creativity
rewarding
emotionally fulfilling
1.5
The goals of interaction design
19
The goals of designing interactive products to be fun, enjoyable, pleasurable,
aesthetically pleasing and so on are concerned primarily with the user experience.
By this we mean what the interaction with the system feels like to the users. This involves explicating the nature of the user experience in subjective terms. For example, a new software package for children to create their own music may be designed
with the primary objectives of being fun and entertaining. Hence, user experience
goals differ from the more objective usability goals in that they are concerned with
how users experience an interactive product from their perspective, rather than assessing how useful or productive a system is from its own perspective. The relationship between the two is shown in Figure 1.7.
Much of the work on enjoyment, fun, etc., has been carried out in the entertainment and computer games industry, which has a vested interest in understanding the role of pleasure in considerable detail. Aspects that have been described
as contributing to pleasure include: attention, pace, play, interactivity, conscious
and unconscious control, engagement, and style of narrative. It has even been
suggested that in these contexts, it might be interesting to build systems that are
non-easy to use, providing opportunities for quite different user experiences from
those designed based on usability goals (Frohlich and Murphy, 1999). Interacting with a virtual representation using a physical device (e.g., banging a plastic
TfUn
----,
emotionally
fulfilling
satisfying
/
efficient
enjoiable
i
entertaining
easy to
remember
how to use
TI
easy to
learn
effective
to use
safe
to use
\
rewarding
1
supportive
of creativity
/
havetgood
utility
\helpful
/
aesthetically
motivating
Figure 1.7 Usability and user experience goals. Usability goals are central to interaction design and are operationalized through specific criteria. User experience goals are shown in
the outer circle and are less clearly defined.
20
Chapter 1
What is interaction design?
I
hammer to hit a virtual nail represented on the computer screen) compared with
using a more efficient way to do the same thing (e.g., selecting an option using command keys) may require more effort but could, conversely, result in a more enjoyable and fun experience.
Recognizing and understanding the trade-offs between usability and user experience goals is important. In particular, this enables designers to become aware of
the consequences of pursuing different combinations of them in relation to fulfilling different users' needs. Obviously, not all of the usability goals and user experience goals apply to every interactive product being developed. Some combinations
will also be incompatible. For example, it may not be possible or desirable to design a process control system that is both safe and fun. As stressed throughout this
chapter, what is important depends on the use context, the task at hand, and who
the intended users are.
elow are a number of proposed interactive products. What do you think are the key usabily goals and user experience goals for each of them?
(a) a mobile device that allows young children to communicate with each other and play
collaborative games
(b) a video and computer conferencing system that allows students to learn at home
(c) an Internet application that allows the general public to access their medical records
via interactive TV
(d) a CAD system for architects and engineers
(e) an online community that provides support for people who have recently been
bereaved
Comment
(a) Such a collaborative device should be easy to use, effective, efficient, easy to learn
and use, fun and entertaining.
(b) Such a learning device should be easy to learn, easy to use, effective, motivating and
rewarding.
(c) Such a personal system needs to be safe, easy to use and remember how to use, efficient and effective.
(d) Such a tool needs to be easy to learn, easy to remember, have good utility, be safe, efficient, effective, support creativity and be aesthetically pleasing.
(e) Such a system needs to be easy to learn, easy to use, motivating, emotionally satisfying and rewarding.
1.6 More on usability: design and usability principles
Another way of conceptualizing usability is in terms of design principles. These are
generalizable abstractions intended to orient designers towards thinking about different aspects of their designs. A well-known example is feedback: systems should
be designed to provide adequate feedback to the users to ensure they know what to
1.6 More on usability: design and usability principles
21
do next in their tasks. Design principles are derived from a mix of theory-based
knowledge, experience, and common sense. They tend to be written in a prescriptive manner, suggesting to designers what to provide and what to avoid at the interface-if you like, the do's and don'ts of interaction design. More specifically, they
are intended to help designers explain and improve the design (Thimbleby, 1990).
However, they are not intended to specify how to design an actual interface (e.g.,
telling the designer how to design a particular icon or how to structure a web portal) but act more like a set of reminders to designers, ensuring that they have provided certain things at the interface.
A number of design principles have been promoted. The best known are concerned with how to determine what users should see and do when carrying out
their tasks using an interactive product. Here we briefly describe the most common
ones: visibility, feedback, constraints, mapping, consistency, and affordances. Each
of these has been written about extensively by Don Norman (1988) in his bestseller
The Design of Everyday Things.
Visibility The importance of visibility is exemplified by our two contrasting exam-
ples at the beginning of the chapter. The voice-mail system made the presence and
number of waiting messages invisible, while the answer machine made both aspects
highly visible. The more visible functions are, the more likely users will be able to
know what to do next. In contrast, when functions are "out of sight," it makes them
more difficult to find and know how to use. Norman (1988) describes the controls
of a car to emphasize this point. The controls for different operations are clearly
visible (e.g., indicators, headlights, horn, hazard warning lights), indicating what
can be done. The relationship between the way the controls have been positioned
in the car and what they do makes it easy for the driver to find the appropriate control for the task at hand.
Feedback Related to the concept of visibility is feedback. This is best illustrated
by an analogy to what everyday life would be like without it. Imagine trying to play
a guitar, slice bread using a knife, or write using a pen if none of the actions produced any effect for several seconds. There would be an unbearable delay before
the music was produced, the bread was cut, or the words appeared on the paper,
making it almost impossible for the person to continue with the next strum, saw, or
stroke.
Feedback is about sending back information about what action has been done
and what has been accomplished, allowing the person to continue with the activity.
Various kinds of feedback are available for interaction design-audio, tactile, verbal, visual, and combinations of these. Deciding which combinations are appropriate for different kinds of activities and interactivities is central. Using feedback in
the right way can also provide the necessary visibility for user interaction.
Constraints The design concept of constraining refers to determining ways of restricting the kind of user interaction that can take place at a given moment. There
are various ways this can be achieved. A common design practice in graphical user
interfaces is to deactivate certain menu options by shading them, thereby restrict-
22
Chapter 1
What is interaction design?
Figure 1.8 A menu illustrating restricted availability of options as an example of logical
constraining. Shaded areas indicate deactivated options.
ing the user to only actions permissible at that stage of the activity (see Figure 1.8).
One of the advantages of this form of constraining is it prevents the user from selecting incorrect options and thereby reduces the chance of making a mistake. The
use of different kinds of graphical representations can also constrain a person's interpretation of a problem or information space. For example, flow chart diagrams
show which objects are related to which, thereby constraining the way the information can be perceived.
Norman (1999) classifies constraints into three categories: physical, logical, and
cultural. Physical constraints refer to the way physical objects restrict the movement of things. For example, the way an external disk can be placed into a disk
drive is physically constrained by its shape and size, so that it can be inserted in
only one way. Likewise, keys on a pad can usually be pressed in only one way.
Logical constraints rely on people's understanding of the way the world works
(cf. the marbles answering machine design). They rely on people's common-sense
reasoning about actions and their consequences. Picking up a physical marble and
placing it in another location on the phone would be expected by most people to
1.6 More on usability: design and usability principles
23
Figure 1.9 (a) Natural mapping between rewind, play, and fast forward on a tape recorder
device. (b) An alternative arbitrary mapping.
trigger something else to happen. Making actions and their effects obvious enables
people to logically deduce what further actions are required. Disabling menu options when not appropriate for the task in hand provides logical constraining. Jt allows users to reason why (or why not) they have been designed this way and what
options are available.
Cultural constraints rely on learned conventions, like the use of red for warning, the use of certain kinds of audio signals for danger, and the use of the smiley
face to represent happy emotions. Most cultural constraints are arbitrary in the
sense that their relationship with what is being represented is abstract, and could
have equally evolved to be represented in another form (e.g., the use of yellow instead of red for warning). Accordingly, they have to be learned. Once learned and
accepted by a cultural group, they become universally accepted conventions. Two
universally accepted interface conventions are the use of windowing for displaying information and the use of icons on the desktop to represent operations and
documents.
Mapping This refers to the relationship between controls and their effects in the
world. Nearly all artifacts need some kind of mapping between controls and effects,
whether it is a flashlight, car, power plant, or cockpit. An example of a good mapping between control and effect is the up and down arrows used to represent the up
and down movement of the cursor, respectively, on a computer keyboard. The
mapping of the relative position of controls and their effects is also important. Consider the various musical playing devices (e.g., MP3, CD player, tape recorder).
How are the controls of playing, rewinding, and fast forward mapped onto the desired effects? They usually follow a common convention of providing a sequence of
buttons, with the play button in the middle, the rewind button on the left and the
fast-forward on the right. This configuration maps directly onto the directionality
of the actions (see Figure 1.9a). Imagine how difficult it would be if the mappings in
Figure 1.9b were used. Look at Figure 1.10 and determine from the various mappings which is good and which would cause problems to the person using it.
Figure 1.10 Four possible combinations of arrow-key mappings. Which is the most natural
mapping?
24
Chapter 1
What is interaction design?
Consistency This refers to designing interfaces to have similar operations and use
similar elements for achieving similar tasks. In particular, a consistent interface is
one that follows rules, such as using the same operation to select all objects. For
example, a consistent operation is using the same input action to highlight any
graphical object at the interface, such as always clicking the left mouse button. Inconsistent interfaces, on the other hand, allow exceptions to a rule. An example of
this is where certain graphical objects (e.g., email messages presented in a table)
can be highlighted only by using the right mouse button, while all other operations
are highlighted using the left button. A problem with this kind of inconsistency is
that it is quite arbitrary, making it difficult for users to remember and making the
users more prone to mistakes.
One of the benefits of consistent interfaces, therefore, is that they are easier to
learn and use. Users have to learn only a single mode of operation that is applicable
to all objects. This principle works well for simple interfaces with limited operations,
like a mini CD player with a small number of operations mapped onto separate buttons. Here, all the user has to do is learn what each button represents and select accordingly. However, it can be more problematic to apply the concept of consistency
to more complex interfaces, especially when many different operations need to be
designed for. For example, consider how to design an interface for an application
that offers hundreds of operations (e.g. a word-processing application). There is
simply not enough space for a thousand buttons, each of which maps onto an individual operation. Even if there were, it would be extremely difficult and timeconsuming for the user to search through them all to find the desired operation.
A much more effective design solution is to create categories of commands
that can be mapped into subsets of operations. For the word-processing application, the hundreds of operations available are categorized into subsets of different
menus. All commands that are concerned with file operations (e.g., save, open,
close) are placed together in the same file menu. Likewise, all commands concerned with formatting text are placed in a format menu. Selecting an operation
then becomes a matter of homing in on the right category (menu) of options and
scanning it for the desired one, rather than scrolling through one long list. However, the consistency rule of having a visible one-to-one mapping between command and operation is broken. Operations are not immediately visible at the
interface, but are instead hidden under different categories of menus. Furthermore,
some menu items are immediately visible, when a top-level menu is first pulled
down, while others remain hidden until the visible items are scrolled over. Thus,
users need to learn what items are visible in each menu category and which are hidden in submenus.
The way the items are divided between the categories of menu items can also
appear inconsistent to users. Various operations appear in menus where they do
not belong. For example, the sorting operation (very useful for listing references or
names in alphabetical order) in Microsoft Word 2001 is in the Table menu (the
Mac Version). In the previous Word 98 version, it was in both the Tools and Table
menus. I always thought of it as a Tool operation (like Word Count), and became
very frustrated to discover that as a default for Word 2001 it is only in the Table
menu. This makes it inconsistent for me in two ways: (i) with the previous version
and (ii) in the category it has been placed. Of course, I can customize the new ver-
1.6 More on usability: design and usability principles
25
sion so that the menus are structured in the way I think they should be, but this all
takes considerable time (especially when I use different machines at work, home,
and when travelling).
Another problem with consistency is determining what aspect of an interface
to make consistent with what else. There are often many choices, some of which
can be inconsistent with other aspects of the interface or ways of carrying out actions. Consider the design problem of developing a mechanism to let users lock
their files on a shared server. Should the designer try to design it to be consistent
with the way people lock things in the outside world (called external consistency)
or with the way they lock objects in the existing system (called internal consistency)? However, there are many different ways of locking objects in the physical
world (e.g., placing in a safe, using a padlock, using a key, using a child safety lock),
just as there are different ways of locking electronically (e.g., using PIN numbers,
passwords, permissions, moving the physical switches on floppy disks). The problem facing designers is knowing which one to be consistent with.
Ahbrdance is a term used to refer to an attribute of an object that allows people
to know how to use it. For example, a mouse button invites pushing (in so doing activating clicking) by the way it is physically constrained in its plastic shell. At a very
simple level, to afford means "to give a clue" (Norman, 1988). When the affordances of a physical object are perceptually obvious it is easy to know how to interact with it. For example, a door handle affords pulling, a cup handle affords
grasping, and a mouse button affords pushing. Norman introduced this concept in
the late '80s in his discussion of the design of everyday objects. Since then, it has
been much popularized, being used to describe how interface objects should be designed so that they make obvious what can be done to them. For example, graphical elements like buttons, icons, links, and scroll bars are talked about with respect
to how to make it appear obvious how they should be used: icons should be designed to afford clicking, scroll bars to afford moving up and down, buttons to afford pushing.
Unfortunately, the term affordance has become rather a catch-all phrase, losing much of its potency as a design principle. Norman (1999), who was largely responsible for originally promoting the concept in his book The Design of Everyday
Things (1988), now despairs at the way it has come to be used in common parlance:
"Zput an affordance there, " a participant would say, "I wonder if the object affords
clicking. . . " affordances this, affordances that. And no data, just opinion. Yikes! What
had I unleashed upon the world? Norman's (1999) reaction to a recent CHI-Web
discussion.
He has since tried to clarify his argument about the utility of the concept by saying
there are two kinds of affordance: perceived and real. Physical objects are said to
have real affordances, like grasping, that are perceptually obvious and do not have to
be learned. In contrast, user interfaces that are screen-based are virtual and do not
have these kinds of real affordances. Using this distinction, he argues that it does not
make sense to try to design for real affordances at the interface--except when designing physical devices, like control consoles, where affordances like pulling and pressing are helpful in guiding the user to know what to do. Alternatively, screen-based
26
Chapter 1
What is interaction design?
interfaces are better conceptualized as perceived affordances, which are essentially
learned conventions. In conclusion, Norman argues that other design concepts--conventions, feedback and cultural and logical constraints-are far more useful for helping designers develop gr...
Purchase answer to see full
attachment