June 4, 1993 17:34

exercises

Sheet number 1 Page number 1

HumanComputer Interaction
Teachers Notes

Alan J. Dix, Janet E. Finlay

University of York, UK

Gregory D. Abowd
Carnegie Mellon University, USA

Russell Beale
University of Birmingham, UK

June 4, 1993 17:34

exercises

Sheet number 2 Page number 2

June 4, 1993 17:34

exercises

Sheet number 3 Page number 1

Introduction
This set of teachers notes accompanies the book HumanComputer Interaction
and is intended to support teachers using the book as a core course text. It includes
solutions to and suggestions for the use of the exercsies in the book, together with
additional project material.
In addition, a set of master slides are available which can be used to produce
overhead projector foils. These cover the key points for each chapter, making them
ideal for a lecture-based course. Please contact the publisher for further details.
We have tried to design the book and additional material to be flexible and to
support effective teaching of HCI. Obviously in any such venture there is room
for improvement. We therefore welcome comments on both the book and this
set of notes which will allow us to improve future editions. We are particularly
keen to hear suggestions for material that you would find useful that we have not
included and, conversely, to know if we have included anything which you do not
find useful. Please send your comments to us care of the publishers.

Course structure
HumanComputer Interaction is a subject which is by definition practical and
whch lends itself well to novel teaching methods. It is our intention that the book
and these notes be able to support both traditional lecture-style courses and those
based mainly on project work.
In either case we strongly recommend that students be given the opportunity
to do some practical work, both in experimenting with and evaluating existing
available systems and in designing their own interfaces. HCI cannot be taught
exclusively through lectures and books and requires some hands-on experience.
These notes offer examples of the types of project that can be used to provide
such experience. As far as possible we have not assumed the availability of any
particular resource so that teachers can adapt them to what is available. However,
if possible, we suggest that all students be given the chance to experiment with
both graphical and command based applications and to use a prototyping system
such as Hypercard or Visual Basic. These allow the student to develop their own
mock up interfaces.
1

June 4, 1993 17:34

exercises

Sheet number 4 Page number 2

Introduction
If such tools are unavailable, drawing packages can be used to design individual screens but have the disadvantage of being static.
For a traditional lecture-based course we suggest the use of the overhead
projector slides (selectively if necessary) backed up by related reading and exercises
from the book and at least one practical design project (see Additional Projects).
If possible other practicals can be included as well. The design project should
encourage the student to use the modelling approaches and to evaluate their
design.
A project-based course can be designed primarily around the exercises, with
suggested recommended related reading. Such a course demands a certain amount
of commitment from students and may therefore be best suited for option courses
and other advanced courses where the students are motivated towards HCI.
Obviously, the choice of exercises and projects will depend largely on the resources and time available but the following is a suggested programme which
assumes 10 2-hour practicals and an individual or group project. In selecting
exercises we have tried to include some that involve research, some that involve observation and some that require practical application. Clearly some of
these can be done in the students own time. If this is the case the extra practical time can be spent introducing different interactive applications and tools.
Week Chapter Exercises
1
1, 2, 3
1.2, 2.3, 3.1 or Project 1
2
4, 5
4.1, 5.1
3
6
6.1, 6.3
4
7
7.5, 7.7
5
8
8.1, 8.2 or Project 2
6
9, 10
9.4, 10.2
7
11
11.1, 11.3, 11.4 or Project 3
8
12
12.2, 12.4
9
13, 14
13.2, 14.2
10
15
15.2, 15.3
At least one design project should be attempted. Depending on time, this can
be done instead of the exercises at the points suggested, or as an end of term project
in addition. Projects 2 and 3 provide most scope for incorporating different aspects
of the course.
An additional resource which we recommend to aid course design is the SIGCHI Curriculum Development Groups report [3]. This provides suggestions for
curricula for different groups of students and for different purposes. The book
covers most of material outlined in the SIGCHI report and can be adapted for use
with most of their suggested curricula.

Exercises and projects

The rest of this booklet contains solutions to exercises from the book and 3 addi-

June 4, 1993 17:34

exercises

Sheet number 5 Page number 3

Introduction
tional design projects which can be used to bring together the techniques from the
different stages of design.
The exercises provided in the book are of three main types: those that require
factual answers (some of which may require additional reading or research), those
that provide practice in using the techniques described, and those which encourage
the student to observe and evaluate existing designs. Consequently the exercises
solutions also differ. For factual questions pointers are given to enable the teacher
to guide the student in the right direction.
However, students may in the process of their research uncover additional
information on a subject and this should be encouraged. In the case of practice
questions example solutions are given. Again variations on these are sometimes
possible (individual solutions indicate this) but the solution given here can be
provided to the student as an example solution.
Observational exercises are usually small practical projects which require the
student to interact with an application or watch someone else do so. Here the
solutions suggest factors that the student should consider in the observation and
hints to the teacher as to how to best encourage effective use of such exercises.
Some exercises refer to sections or diagrams in the book itself, others to diagrams internal to these notes. The former are referred to using the same notation
as the book; the latter are distinguished by the use of the letter x i.e. Figure 12.1
refers to Figure 12.1 in the book while Figure 12.x1 refers to a Figure in the notes.
The three additional projects are more extensive than any of the exercises and
can be used as end of term projects or assessments. They are not included in the
book itself for this reason.
They require knowledge of more than one aspect of design and are intended
to demonstrate how the different stages of design fit together. Each project is
described in full, together with any support material, and hints to the teacher on
how to use the projects are given. These projects can be reproduced for the purposes
of teaching on courses where the core text is HumanComputer Interaction.

June 4, 1993 17:34

exercises

Sheet number 6 Page number 4

June 4, 1993 17:34

exercises

Sheet number 7 Page number 5

Chapter 1

The Human

Exercise 1.1
Suggest ideas for an interface which uses the properties of sound effectively.

Answer
This is an opportunity for the student both to use his/her imagination and/or to
do a literature survey (starting with the references in Chapter 15). One possibility
for tackling the exercise is to encourage the student to think how sound could be
added to an application with which s/he is familiar.
Speech sounds can obviously be used to convey information. This is useful
not only for the visually impaired but also for any application where the users
attention has to be divided (for example, power plant control, flight control etc.).
Uses of non-speech sounds include






Attention to attract the users attention to a critical situation or to the end

of a process for example.
Status information continuous background sounds can be used to convey
status information. For example, monitoring the progress of a process (without
the need for visual attention).
Confirmation a sound associated with an action to confirm that the action
has been carried out. For example, associating a sound with deleting a file.
Navigation using changing sound to indicate where the user is in a system.
For example, what about sound to support navigation in hypertext?

June 4, 1993 17:34

exercises

Sheet number 8 Page number 6

The Human

Exercise 1.2
Devise experiments to test the properties of (i) short-term memory (ii)
long-term memory, using the experiments described in this chapter to
help you. Try out your experiments on your friends. Are your results
consistent with the properties described in this chapter?

Answer
The aim of this exercise is to get the student to think about experimental design.
The experiments devised can effectively be repeats of the originals. Chapter 11 can
be used for reference on experimental design.

(i) Short term memory (STM)

The student should first choose an aspect to investigate, for example, digit span,
recency effect, decay.

Example solution: STM decay

Subjects ideally selected to represent population, more probably undergraduate
students (try to get a range of academic subjects). Sample size: 10+
Experiment split subjects into two groups. Each subject studies list of 15-20 words
(could try with both nonsense words and actual words to see any difference).
Subject has to recall list either (a) immediately or (b) after 20 second delay.
Measure the number (or percentage ) of the words remembered correctly. A
within groups design can be used to avoid individual bias or group variation
(as long as different lists are used for each attempt).
independent variable delay in recall
dependent variable number correctly recalled.
Group (b) should be given a task to do during the delay period in order to
avoid rehearsal. If possible this task should occupy a different channel to
minimise interference, e.g., a visual recognition task.
Hypothesis Those in (b) will perform worse than those in (a) since STM will decay.
Analysis graphs to see decay. T test

June 4, 1993 17:34

exercises

Sheet number 9 Page number 7

The Human

(ii) Long term memory (LTM)

The student should first choose an aspect to investigate, for example, the total time
hypothesis or the distribution of practice effect.

Example solution: distribution of practice effect

Subjects as above. Should have no prior experience of the skill to be learned.

Experiment split subjects into 3 groups. Each subject must learn a skill (for example short hand or typing must be measurable). Group A learns for 1 hour
a week for 6 weeks. Group B learns for 2 hours a week for 3 weeks. Group C
learns for 3 hours a week for 2 weeks. After each groups training is complete
the subjects are tested and the number of mistakes made noted.
independent variable style of learning
dependent variable accuracy
Between groups design.
Hypothesis Group A will be best (due to the distribution of practice effect)
Analysis ANOVA
(N.B. This one is not easy to run but could be done with cooperation from friends)

Exercise 1.3
Identify the goals and operators involved in the problem delete the second
paragraph of the document on a word-processor. Now use a wordprocessor to delete a paragraph and note your actions, goals and sub-goals.
How well did they match your earlier description?

Answer
Assume you have a document open and you are at some arbitrary position within it.
You also need to decide which operators are available and what their preconditions
and results are. Based on an imaginary word processor we assume the following
operators (you may wish to use your own WP package):

June 4, 1993 17:34

exercises

Sheet number 10 Page number 8

The Human
Operator

Precondition

Result

delete paragraph

cursor at start
of paragraph
cursor anywhere
in document

paragraph
deleted
cursor moves
to start of next
paragraph (except
where there is no
next paragraph when
no effect)
cursor at start of
document

move to paragraph

move to start

cursor anywhere
in document

Goal: delete second paragraph in document

Looking at the operators an obvious one to resolve this goal is delete paragraph
which has the pre-condition cursor at start of paragraph. We therefore have a new
sub-goal: move to paragraph. The pre-condition is cursor anywhere in document
(which we can meet) but we want the second paragraph so must initially be in the
first.
We set up a new subgoal, move to start, with pre-condition cursor anywhere in
document and result cursor at start of document. We can then apply move to paragraph
and finally delete paragraph.
We assume some knowledge here (that the second paragraph is the paragraph
after the first one).

Exercise 1.4

Observe skilled and novice operators in a familiar domain, for example,

touch and hunt-and-peck typists, expert and novice game players, or
expert and novice users of a computer application. What differences can
you discern between their behaviours?

Answer
This is an exercise in observation. The student should think about skill acquisition,
proceduralization, chunking etc. Is there any evidence of this in practice? How do
the groups differ (speed, error, style, strategy)? Do the differences suggest different
skill levels.

June 4, 1993 17:34

exercises

Sheet number 11 Page number 9

The Human
STM

Skill acquisition
Mental models

Memory
LTM

Deductive

Thinking

Reasoning

Sensory
Abductive

Problem solving
Iconic

Inductive

Echoic
Gestalt

Vision

Hearing

Cognition

Analogy

Problem space

Perception
Psychology

Figure 1.x1 The top-down view

Exercise 1.5
Produce a semantic network of the main information in this chapter.

Answer
This network is potentially huge so it is probably unnecessary to devise the whole
thing! Encourage the student to be selective. One helpful way to tackle the exercise
is to approach it in both a top down and a bottom up manner. Top down will give
you a general overview of topics and how they relate; bottom up can fill in the
details of a particular field. These can then be glued together to build up the
whole picture. Perhaps a group of students could tackle the problem together,
each taking one part of it. We will not provide the full network here but will
give examples of the level of detail anticipated for the overview and the detailed
versions. In the overview we have not included labels on the arcs for clarity.

June 4, 1993 17:34

10

exercises

Sheet number 12 Page number 10

The Human

digit span

chunking
is
increased by

STM

subject to

has

limited capacity
shown by

recency effect
Figure 1.x2 The bottom-up view

shown by

Miller

June 4, 1993 17:34

exercises

Sheet number 13 Page number 11

Chapter 2

The Computer

The aim of this chapter is to give the students a general view of the capabilities of
typical computer systems insofar as they impact upon the users. This is obviously
particularly relevant for students from a human sciences background as they may
be unaware of the basic components of computer equipment. However, many
computer scientists are equally ignorant when it comes to the gross capabilities of
standard systems. They may be able to tell you that a bubble sort is an order n2
algorithm whereas quicksort is order n log n, but may have no idea how long a
typical database would take to sort 1000 records!

Exercise 2.1

What is the basic architecture of a computer system?

Answer
Some students might have done a computer architectures course, but they should
not give the stock answer for such a question in that context. Instead, they should
be assessing the architecture from the point of view of the user. The material for
this question is scattered throughout the chapter. However, students should also
be directed to personal computer magazines where adverts and articles will give
them some idea of typical capabilities and costs. It might also prompt some
questions: just what is the difference to the user between a 16ms and a 20ms disk
drive?
The example answer below gives the general style, although more detail would
be expected of a full answer. In particular, the students should be encouraged to
have a feel for capacities either as ball park figures or in terms of typical capabilities
(seconds of video, pages of text).
11

June 4, 1993 17:34

12

exercises

Sheet number 14 Page number 12

The Computer
Example
The basic architecture of a computer systems consists of the computer itself (with
associated memory), input and output devices for user interaction and various
forms of hard copy devices. (Note, the computer science answer regards output to
the user and output to a printer as essentially equivalent. This is not an acceptable
user centred view.)
A typical configuration of user input/output devices would be a screen with a
keyboard for typing text and a mouse for pointing and positioning. Depending on
circumstance, different pointing devices may be used such as lightpen (for more
direct interaction) or a trackball (especially on portable computers).
The computer itself can be considered as composed of some processing element
and memory. The memory is itself divided into short term memory which is lost
when the machine is turned off and permanent memory which persists.

Exercise 2.2
How do you think new, fast, high-density memory devices and quick
processors have influenced recent developments in HCI? Do they make
systems any easier to us? Do they expand the range of applications of
computer systems?

Answer
Arguably it is not so much the increase in computer power as the decrease in the
cost of that power which has had the most profound effect. Because, ordinary
users have powerful machines on their desktops it has become possible to view
that power as available for the interface rather than hoarded for number crunching
applications.
Modern graphical interaction consumes vast amounts of processing power and
would have been completely impossible only a few years ago. There is an extent
to which systems have to run faster to stay still, in that as screen size, resolution
and colour range increase, so does the necessary processing power to maintain the
same interaction. However, this extra processing is not really producing the same
effect, screen quality is still a major block on effective interaction.
The increase in RAM means that larger programs can be written, effectively
allowing the programmer elbow room. This is used in two ways: to allow extra
functionality and to support easier interaction. Whether the former really improves
usability is debatable unused functionality is a good marketing point, but is of
no benefit to the user. The ease of use of a system is often determined by a host
of small features, such as the appropriate choice of default options. These features
make the interface seem simple, but make the program very complex and large.

June 4, 1993 17:34

exercises

Sheet number 15 Page number 13

The Computer
Certainly the availability of elbow room, both in terms of memory and processing
power, has made such features possible.
The increase in both short term (RAM) and long term (disks and optical storage)
has also removed many of the arbitrary limits in systems: it is possible to edit
documents of virtually unlimited size and to treat the computer (suitably backed
up) as ones primary information repository.
Some whole new application areas have become possible because of advances
in memory and processing. For example, most applications of multi-media, for
example voice recognition and online storage and capture of video and audio,
require enormous amounts of processing and/or memory. In particular, large
optical storage devices have been the key to electronic document storage whereby
all paper documents are scanned and stored within a computer system. In some
contexts such systems have completely replaced paper based filing cabinets.

Exercise 2.3
What input and output devices would you use for the following systems?
For each, compare and contrast alternatives, and if appropriate indicate
why the conventional keyboard, mouse and c.r.t screen may be less suitable.
a) portable word processor
b) tourist information system
c) tractor-mounted crop-spraying controller
d) air traffic control system
e) worldwide personal communications system
f) digital cartographic system

Answer
Whereas question 2.1 focuses on typical systems, the emphasis here is on the
diversity of different devices needed for specialised purposes. During classes
and practicals the students can be exposed to a wide variety of input and output
devices. They can also be encouraged to collect devices, that is, to watch out for
shop tills, bank tellers, taxi meters, lift buttons, domestic appliances etc.
a) portable word processor
The determining factors are size, weight and battery power. However, remember the purpose, this is a word processor not an address book or even data
entry device.
LCD screen low power requirement

13

June 4, 1993 17:34

14

exercises

Sheet number 16 Page number 14

The Computer
Trackball or stylus for pointing
Real keyboard you cant word process without a reasonable keyboard
and stylus handwriting recognition is not good enough.
Small, low power bubble-jet printer although not always necessary,
this makes the package stand alone. It is probably not so necessary that
the printer have large battery capacity as printing can probably wait until
a power point is found.
b) tourist information system
This is likely to be in a public place. Most users will only visit the system once,
so the information and mode of interaction must be immediately obvious.
Touch screen only easy and direct interaction for first time users (see
also Chapter 3).
NO mice or styluses in a public place they wouldnt stay long!
c) tractor-mounted crop-spraying controller.
A hostile environment with plenty of mud and chemicals. Requires numerical
input for flow rates etc., but probably no text.
Touch sensitive keypad , ordinary keypads would get gunged up.
Small dedicated LED display (LCD often cant be read in sunlight and
large screens are fragile.
Again no mice or styluses they would get lost.
d) air traffic control system
The emphasis is on immediately available information and rapid interaction.
The controller cannot afford to spend time searching for information, all frequently used information must be readily available.
Several specialised displays including overlays of electronic information on radar.
Lightpen or stylus high precision direct interaction.
Keyboard for occasional text input, but consider making it fold out of
the way.
e) worldwide personal communications system
Basically a super mobile phone! If is to be kept on hand all the time it must be
very light and pocket sized. However, to be a communications system one
would imagine that it should also act as a personnel address/telephone book
etc.
Standard telephone keypad the most frequent use
Small dedicated LCD display low power, specialised functions.

June 4, 1993 17:34

exercises

Sheet number 17 Page number 15

The Computer
Possibly stylus for interaction it allows relatively relatively rich interaction with the address book software, but little space.
A docking facility the system itself will be too small for a full sized
keyboard (!), but you wont want to enter in all your addresses and telephone numbers by stylus!
f) digital cartographic system
This calls for very high precision input and output facilities. It is similar to
CAD in terms of the screen facilities and printing, but in addition will require
specialised data capture.
Large high resolution colour VDU (20 inch or bigger) these tend to be
enormously big (from back to front). LCD screens, although promising
far thinner displays in the long term, cannot at present be made large
enough.
Digitising tablet for tracing data on existing paper maps. It could also
double up as a pointing device for some interaction.
Possibly thumbwheels for detailed pointing and positioning tasks.
Large format printer indeed VERY large an A2 or A1 plotter at minimum.

15

June 4, 1993 17:34

exercises

Sheet number 18 Page number 16

June 4, 1993 17:34

exercises

Sheet number 19 Page number 17

Chapter 3

The Interaction

Exercise 3.1
Choose two of the interaction styles (described in Section 3.3) that you
have experience of using. Use the interaction framework to analyze the
interaction involved in using these interface styles for a database selection
task. Which of the distances is greatest in each case?

Answer
There is no single answer for this exercise, so we will provide an example of the
style of answer that is suitable and the level of analysis which is appropriate.
The students should be aware that although the term distance is used, we have not
associated any real measures to any of the translations in the interaction framework.
As a result, this analysis can only be informal and at this point is mainly informed
by the students intuition and experience with various interaction styles.
As was stated in Section 3.2.3, assessment of any interaction with the interaction
framework can only be relative to some task. For this example we will choose
a common database selection task that of selecting records from an online
library database. The two interaction styles we will analyze are a natural language
interface and a command line interface.
The task is to select a set of references from the library database that satisfy
some search criteria. Once the task has been formulated in the users task language
(for instance, the user wants to see all of the books written by Alan Dix since 1990),
that task must be articulated in the input language. A natural language interaction
style would allow the user simply to type in the selection query exactly as they
think of it. The articulation distance is small because it is both easy to articulate
(possibly even easier if a spoken interface is provided rather than typing) and the
coverage is total (the user is allowed to articulate anything as a query). On the other
hand, for a command line interface, the limited vocabulary of the input language
makes it more difficult for the user to articulate a task even though the limited
language provides complete coverage in terms of possible queries allowed. The
17

June 4, 1993 17:34

18

exercises

Sheet number 20 Page number 18

The Interaction
real difficulty for a natural language interface is how the system translates the input
expression into the actual query that accesses the library records. This performance
translation would be much easier for the command line interface since it may not
even require any translation of an input expression, that language having already
been constructed with the database engine in mind. Therefore, we can see that for
a natural language interface, the performance distance is greatest, whereas for a
command line interface it is the articulation distance which is greatest.
But the above analysis only really deals with the execution translations. On
the evaluation side, a natural language interface must try to present the results
of the database query in the form in which the user phrased the question. This
could in general be a difficult translation for the system as it attempts to answer
questions in the style in which an arbitrary user has posed that question. Having
accomplished that, the observation by the user should be easy to perform. For
a command line interface, there is no guarantee that the result of the query will
be automatically displayed and the user may have to explicitly request a display
(and they may have to express how the display be formatted). Neglecting that
point, presentation by the system is made easier as the output language can be
very constrained. Observation is made more difficult as the user must translate
the output into the terms of their original task formulation. For example, having
asked for books by Alan Dix published after 1990, the user may have a difficult time
locating author name and year of publication to determine if the resulting records
match their expectations. For evaluation, a natural language interface has a greater
presentation distance and a command line interface a greater observation distance.
In general, therefore, we would expect that a natural language interface would be
easier from the users perspective but more difficult from the system builders
perspective. The opposite should hold for a command language interface.
There are some issues that we havent addressed in this example, such as
displaying a large set of records that satisfy the query and being able to reuse the
output of one query as the input to another to easily compound searches. Since
the performance translation is so difficult for a natural language interaction style it
is important for a natural language interface to present the results of the query in
such a way that the user is able to determine if the system understood the original
query in the way the user intended. This would involve the presentation translation
both reiterating the users query and the selected records simultaneously. In our
example, since the user was interested in the author and date of publication (Dix,
after 1990), it would help if that information was prominently presented in the
result set.
We also have not considered what effect experience with the system provides.
As users become more comfortable with the syntax and semantics of a command
language, its perceived difficulty will decrease. Also, a verbose natural language
output could limit the number of records from a result set that could be displayed.
The moral of the story is that despite their intuitive allure, such informal analyses
as suggested by this exercise cannot be the last word on analysis of an interactive
system. Ultimately, our judgements must be made more precise and concrete.

June 4, 1993 17:34

exercises

Sheet number 21 Page number 19

The Interaction

Exercise 3.2
Find out all that you can about natural language interfaces. Are there any
successful systems? For what applications are these most appropriate?

Answer
This exercise is intended to encourage the student to do some personal research in
the form of a brief literature survey. Pointers can be given to appropriate places
to begin such a search. For example, general text books on Artificial Intelligence
will include basic information on natural language processing and famous systems
such as SHRDLU; proceedings from conferences such as the AAAI and specialist
journals will have more up to date research papers in the area; proceedings of
HCI conferences such as CHI, Interact and HCI will include natural language
systems particularly geared towards the interface. Other likely sources are popular
journals such as BYTE and personal computer magazines which are likely to review
commercial systems. The students response to the second part of the question will
depend upon what is unearthed, but it is likely that the systems that they find out
about operate in very constrained domains and that the natural language used is
restricted. There are as yet no general purpose natural language interfaces.

Exercise 3.3
What influence does the social environment in which you work have on
your interaction with the computer? What effect does the organization
(commercial or academic) to which you belong have on the interaction?

Answer
The aim of this exercise is to get the student to explore the social and environmental
influences which effect interaction, often without the user being aware of them. The
particular influences will vary from environment to environment but the student
should be encouraged to consider some or all of the following.






work context is the work place shared? are the machines shared?
peer pressure is there pressure to compete or impress?
management pressure is there pressure to achieve? Is the interaction carried
out in the presence of management?
motivation - what motivates the interaction? Does this encourage or discourage experimentation?

19

June 4, 1993 17:34

20

exercises

Sheet number 22 Page number 20

The Interaction




organizational goals what is the objective of the organization? (profit?

education? etc.) How does this effect the interaction?
organizational decision making - who determines the systems that you use?
Do you have any choice or influence? Does this influence the way you interact
with the system?

In each case the student should discuss what influence this may have on the
interaction. It may be helpful to consider other possible environments in order
to identify how the interaction would differ under these different circumstances.
For example, if the student currently shares a machine with colleagues, would
his/her interaction practice change if s/he was given a private machine? Chapter
14 also discusses the influence of groups of workers within an organization on the
an interaction, and is suggested as further reading material on this topic.

June 4, 1993 17:34

exercises

Sheet number 23 Page number 21

Chapter 4

Usability Paradigms and Principles

Exercise 4.1
Look up and report back guidelines for the use of colour. Be able to state
the empirical psychological evidence which supports the guidelines. Do
the guidelines conflict with any other known guidelines? Which principles
of interaction do they support?

Answer
(Sample) There are many examples of guidelines for the use of colour in the
literature. Here are three good sources:





Brown, C. Marlin, Human-Computer Interface Design Guidelines, Ablex, 1988.

Mayhew, Deborah J., Principles and Guidelines in Software User Interface Design,
Prentice-Hall, 1992.
Sun Microsystems, Inc., OPEN LOOK Graphical User Interface Application Style
Guidelines, Addison-Wesley, 1990.

Taking an example from Mayhew, we have the following design guideline for the
use of colour as an informational cue for the user (e.g., to inform the user that a
string of text is a warning or error message):
Do not use colo(u)r without some other redundant cue
Mayhew provides 3 reasons which empirically support this guideline:
1. Colour may not be available on all machines on which the system is to be
implemented. Therefore, if use of colour is the only means to convey some
important information to the user, then that information will be lost in a monochrome (no colour) system. Redundant colour coding will allow for portability
across different computing platforms.
21

June 4, 1993 17:34

22

exercises

Sheet number 24 Page number 22

Usability Paradigms and Principles

2. Empirical evidence shows that 8% of the (general) male population and 0.4%
of the female population has some colour deficiency, so they cannot accurately
recognize or distinguish between various colours. Again, if colour is the only
means for conveying some information, this significant portion of the user
population will be slighted.
3. It has been shown that redundant colour coding enhances user performance
This guideline supports several of the principles discussed in this chapter:
Substitutivity The system is able to substitute colour-coded information and other
means (e.g., text, sound) to represent some important information. We could
turn the argument around and suggest that then the user be able to provide
colour input (by selecting from a palette menu) or other forms of input to
provide relevant information to the system.
Observability This principle is all about the system being able to provide the
user with enough information about its internal state to assist the users task.
Relying strictly on colour-coded information, as pointed out above, could
reduce the observability of a system for some users.
Synthesis If a change in colour is used to indicate the changing status of some
system entity (perhaps a change in temperature above a threshold value is
signaled by an icon becoming red), those who cannot detect the change in
colour would be deprived of this information. Synthesis is about supporting
the users ability to detect such significant changes, especially when they are
a result of previous user actions.
There is no evidence of existing guidelines which this particular guideline for
colour violates.
Another example of a colour guideline (found in all three of the above references) is the demand to consider cultural information in the selection of particular
colours. For example, Mayhew states that Western cultures tend to interpret green
to mean go or safe, red to mean stop, on, hot or emergency and blue to mean cold or
off. Using colour to suggest these kinds of meanings is in support of the familiarity
principle within learnability. However, in other cultures different meanings may
be associated with these colours and consistent use of colour (another guideline)
might lead to confusion. Hence, strict adherence to this guideline would suggest
a violation of the consistency of colour application guideline. However, if consistency is applied relative to the meaning of the colour (as opposed to its actual
colour), this guideline would not have to conflict.

June 4, 1993 17:34

exercises

Sheet number 25 Page number 23

Usability Paradigms and Principles

Exercise 4.2
What was the problem with the synthesis example comparing a command
language interface with a visual interface? Can you suggest a fix to make
a visual interface really immediately honest?

Answer
In this example to demonstrate the principle of synthesizability within learnability,
it was stated that a visual interface to a file management system provided immediate information about the changed location of some file after a move operation
performed by the user. In contrast, a command language interface requires the
user to remember the directory to which a file was moved and explicitly issue
commands to browse the directory to verify that the file has been moved. To really
be sure that a move occurred, the user would have to also browse the original
directory to determine that the file is no longer there. The fallacy in this argument
is that visual file management systems do always provide information about the
new whereabouts of a moved file. To take a counterexample using the Macintosh
example from the text, if a file is moved from one open folder (in which the contents
of the folder is revealed to the user) to a closed folder (contents not revealed) then
the location of the moved file is not indicated to the user unless she remembers to
open up the destination folder to reveal its contents. This is an example of eventual
honesty and not immediate, as the example suggests.
We could fix this problem of eventual honesty for the visual system by
demanding that the destination folder be an open folder (probably too restrictive,
given the limited screen size) or by having the destination folder temporarily open
up to reveal that the file is now located within it. This last suggestion is also a bit
tricky, for we would still want to determine that the file no longer resides in the
original folder, so we would have to be sure that the new folder does not obstruct
the view of the old folder. In practice, this might be too difficult to guarantee in
general.

Exercise 4.3

It has been suggested in this chapter that consistency could be considered

a major category of interactive principles, on the same level as learnability,
flexibility and robustness. If this had been the case, which principles
discussed in this chapter would appear in support of consistency?

23

June 4, 1993 17:34

24

exercises

Sheet number 26 Page number 24

Usability Paradigms and Principles

Answer
As mentioned in the discussion of consistency, it can take many forms because
consistency is usually referred to relative to some other feature of the interaction
between user and system. Mentioned already in the text we have consistency
related to the following principles:
Familiarity consistency with respect to prior real-world experience
Generalizability consistency with respect to experience with the same system or
set of applications on the same platform
In addition, we could interpret some other principles as contributors to consistency:
Affordance consistency with understood intrinsic properties of an object, so a soft
button on the screen should allow us to always push on it to select some
action
Predictability consistency of system response with users expectation, given the
user has some information about past interaction history
Substitutivity consistent permission from system to allow use of equivalent values for input and output
Commensurate effort consistency of effort with respect to doing and undoing
tasks
Response time stability consistency of system response for similar actions
Some other principles for consistency from the text and elsewhere:

Consistency can be relative to the form of input/output expressions relative to

users conceptual model of the system. An example in the text involve using
keys whose relative positions are similar to commands for the systems (any
set of four typewriter keys which form a diagonal to indicate up, down, left
and right information for an input command).
As discussed in the exercise on colour, consistency can be with respect to social
or cultural conventions (e.g., using red to indicate stop or hot, green for go,
blue for cool).

June 4, 1993 17:34

exercises

Sheet number 27 Page number 25

Usability Paradigms and Principles

Exercise 4.4
Discuss the ways in which a full-page word-processor is or is not a direct
manipulation interface for editing a document using Shneidermans criteria. What features of a modern word processor break the metaphor of
composition with pen (or typewriter) and paper?

Answer
We will answer the first question by evaluating the word-processors relative to the
criteria for direct manipulation given by Shneiderman.

Visibility of the objects of interest The most important objects of interest in a

word-processor are the words themselves. Indeed, the visibility of the text
on a continual basis was one of the major usability advances in moving from
line-oriented to display-oriented editors. Depending on the users application,
there may be other objects of interest in word-processing that may or may not
be visible. For example, are the margins for the text on screen similar to the
ones which would eventually printed? Is the spacing within a line and the
line-breaks similar? Are the different fonts and formatting characteristics of
the text visible (without altering the spacing)? Expressed in this way, we can
see the visibility criterion for direct manipulation as very similar to the criteria
for a WYSIWYG (What You See Is What You Get) interface.
incremental action at the interface with rapid feedback on all actions We expect from
a modern word-processor that characters appear in the text as we type them it
at the keyboard, with little delay. If we are inserting text within a paragraph,
we might also expect that the format of the paragraph adjust immediately to
accommodate the new changes. Various word processors do this reformatting
automatically, whereas others do it occasionally or only at the explicit request
of the user. One of the other important actions which requires incremental and
rapid feedback is movement of the insertion point, usually by means of arrow
keys. If there is a significant delay between the input command to move the
insertion point down one line and the actual movement of the cursor on screen,
it is quite possible that the user will overshoot the target when repeatedly
pressing the down-arrow key to move down a few lines on the screen.
Reversibility of all actions, so that users are encouraged to explore without severe
penalties Single step undo commands in most word-processors allow the user
to recover from the last action performed. One problem with this is that the user
must recognize the error before doing any other action. More sophisticated
undo facilities allow the user to retrace back more than one command at a
time. The kind of exploration this reversibility provides in a word-processor
is best evidenced with the ease of experimentation that is now available for

25

June 4, 1993 17:34

26

exercises

Sheet number 28 Page number 26

Usability Paradigms and Principles

formatting changes in a document (fonts types and sizes and margin changes).
One problem with the ease of exploration is that emphasis may move to the
look of a document rather than what the text actually says (style over content).

Syntactic correctness of all actions, so that every operation is a legal operation WYSYWYG word-processors usually provide menus and buttons which the user
uses to articulate many commands. These interaction mechanisms serve to
constrain the input language to only allow legal input from the user. Document production systems, such as troff, TeX, and Scribe, force the user to input
textual commands (which may be erroneously entered by the user) to achieve
desired formatting effects.
Replacement of complex command languages with actions to manipulate directly the
visible objects The case for word processors is similar to that described above for
syntactic correctness. In addition, operations on portions of text are achieved
many times by allowing the user to directly highlight the text with a mouse
(or arrow keys). Subsequent action on that text, such as moving it or copying
it to somewhere else, can then be achieved more directly by allowing the user
to drag the selected via the mouse to its new location.

To answer the second question concerning the drawback of the pen (or typewriter) metaphor for word-processing, we refer to the discussion on metaphors in
Section 4.2.6. The example there compares the functionality of the space key in
typewriting versus word-processing. For a typewriter, the space key is passive, it
merely moves the insertion point one space to the right. In a word processor, the
space key is active, as it inserts a character (the space character) into the document.
The functionality of the typewriter space key is produced by the movement keys
for the word-processor (typically an arrow key pointing right to move forward
within one line). In fact, much of the functionality that we have come to expect of a
word-processor is radically different from that expected of a typewriter, so much so
that the typewriter as a metaphor for word-processing is not all that instructive. In
practice, modern typewriters have begun to borrow from word-processors when
defining their functionality!

June 4, 1993 17:34

exercises

Sheet number 29 Page number 27

Chapter 5

The Design Process

Exercise 5.1
Starting with some of the principles outlined in Chapter 4, provide a
usability specification for an electronic meetings diary or calendar. First
identify some of the tasks that would be performed by a user trying to
keep track of future meetings, and then complete the usability specification
assuming that the electronic system will be replacing a paper-based system.
What assumptions do you have to make about the user and the electronic
diary in order to create a reasonable usability specification?

Answer
This exercise could be easily extended to a small project which would involve
the design of such an electronic diary or calendar. The purpose of this smaller
usability engineering exercise is to show how usability goals can be formulated
early on to drive the design activity. We will select two of the usability principles
from Chapter 4 which will serve as attributes for separate usability specifications.
In the first example, we will consider the interaction principle of guessability,
which concerns how easy it is for new users to perform tasks initially. The measuring concept will be how long it takes a new user, without any instruction on
the new system, to enter their first appointment in the diary. A sample usability
specification is given below.
Attribute: Guessability
Measuring Concept: Ease of first use of system without training
Measuring Method: Time to create first entry in diary
Now Level: 30 seconds on paper-based system
Worst Case: 1 minute
Planned Level: 45 seconds
Best Case: 30 seconds (equivalent to now)
The values in this usability specification might seem a little surprising at first,
since we are saying that the best case is only equivalent to the currently achievable
27

June 4, 1993 17:34

28

exercises

Sheet number 30 Page number 28

The Design Process

level now. The point in this example is that the new system is replacing a very
familiar paper and pencil system which requires very little training. The objective
of this system is not so much to improve guessability but to preserve it. In the
chapter, we discussed that the worst case level should not usually be worse than
the now level, but we are hoping for this product to improve overall functionality
of the system. The user will be able to do more things with the electronic diary
than they could with the conventional system. As a result, we worry less about
improving its guessability. Perhaps we could have been more ambitious in setting
the best case value by considering the potential for voice input or other exotic input
techniques which would make entry faster than writing.
As another example, this time within the flexibility category, we want to support the task migratability of the system. A frequent sort of task for a diary is to
schedule weekly meetings. The conventional system would require the user to
make an explicit entry for the meeting each weekthe task of the scheduling is the
responsibility of the user. In the new system, we want to allow the user to push the
responsibility of scheduling over to the system, so that the user need only indicate
the desire to have a meeting scheduled for a certain time each week and the system
will take care of entering the meeting at all of the appropriate times. The task of
scheduling has thus migrated over to the system. The usability specification for
this example follows.
Attribute:
Measuring Concept:
Measuring Method:

Task migratability
Scheduling a weekly meeting
Time it takes to enter a weekly meeting
appointment
Now Level: (time to schedule one appointment)  (number of
weeks)
Worst Case: time to schedule two appointments
Planned Level: 1.5  (time to schedule one appointment)
Best Case: time to schedule one appointment
In this specification, we have indicated that the now level is equivalent to the
time it takes to schedule each appointment separately. The worst, planned and best
case levels are all targeted at some proportion of the time it takes to schedule just a
single appointmenta dramatic improvement. The difference between the worst,
planned and best case levels is the amount of overhead it will take to indicated
that a single appointment is to be considered an example to repeated at the weekly
level.
What are the assumptions we have to make in order to arrive at such a usability
specification? One of the problems with usability specifications, as we have stated
in the chapter, is that they sometimes require quite specific information about the
design in order to be expressed. For example, had we set one of our measuring
methods to count keystrokes or mouse clicks, we would have had to start making
assumptions about the method of interaction that the system would allow. Had
we tried to set a usability specification concerning the browsing of the diary, we
would have had to start making assumptions about the layout of the calendar

June 4, 1993 17:34

exercises

Sheet number 31 Page number 29

The Design Process

(monthly, weekly, daily) in order to make our estimates specific enough to measure.
In the examples we have provided above, we have tried to stay as abstract as
possible, so that the usability specifications could be of use as early in the design
life cycle as possible. A consequence of this abstractness, particularly evident in
the second example, is that we run the risk in the usability specification of setting
goals that may be completely unrealistic, though well-intentioned. If the usability
specification were to be used as a contract with the customer, such speculation
could spell real trouble for the designer.

Exercise 5.2
Can you think of any instances in which the nounverb guideline for
operations, as suggested in the Apple human interface guidelines for the
Desktop Interface, would be violated? Suggest other abstract guidelines or
principles besides consistency which support your example. (Hint: Think
about moving files around on the Desktop.)

Answer
The noun-verb guideline suggests that we can view all operations that the user will
perform as being composed of an action (the verb) acting with one argument (the
noun). In the case of moving a file (or copying for that matter), the action (move or
copy) requires more than one argument. The way the move operation is performed
requires the user to first select the icon for the file to be moved and then indicate the
move operation implicitly by dragging the selected icon to the destination folder.
The nouns in this dialogue are the file to be moved and the destination folder.
The verb is the move operation. The natural way to express this is in the order
noun-verb-noun. Strictly speaking, in order to stick with the noun-verb guideline,
we would have to indicate both the target file and the destination folder before
indicating the move operation. That would be consistent with relative to input
expression with most other commands on the desktop. However, some principles
of direct manipulation and the familiarity are more important. Moving files by
dragging them on the desktop is very similar to the way we can pick up any object
in the physical world and move it to its new location. And the dragging operation
is incremental and easily recoverable; moving to one place can be undone within
the same operation since the dragging can continue until the file is released.
The file moving example is a slightly contrived one, because some could argue
that there is no violation of the noun-verb guideline (hence, moving is still consistent with respect to input expression) because the verb is move to destination
folder. Perhaps a better example is a command to search a file system for files
matching some specification. Here, the action is to do the qualified search and
the argument or noun is the set of folders or volumes of the system that you want
searched. Typically, this kind of operation is defined by some dialogue box that

29

June 4, 1993 17:34

30

exercises

Sheet number 32 Page number 30

The Design Process

allows the user to indicate in any order the specifics of the operation (the search
parameters) and the folders or volumes to search. Once this unordered dialogue
is complete, the user then indicates that it is OK for the system to perform the
operation. This kind of form-filling dialogue prescribes to neither the noun-verb
or verb-noun guideline; the order is more flexible for the user than consistent.

Exercise 5.3
Can you think of any instances in which the user control guideline suggested by Apple is not followed? (Hint: Think about the use of dialogue
boxes.)

Answer
The user control guideline states that, The user, not the computer, initiates and
controls all actions. In the case of dialogue boxes, this guideline is clearly contradicted. A dialogue box can be used to indicate when an error occurs in the system.
Once this error has been detected and presented to the user in the dialogue box, the
only action that the system allows the user is to acknowledge the error and dismiss
the dialogue box. The system preempts the user dialogue, with good reason. The
preemptive nature of the dialogue box is to ensure that the user actually notices
that there was an error. Presumably, the only error that will be produced in such an
intrusive manner are ones which the user must know about before proceeding, so
the preemption is warranted. But sometimes dialogue boxes are not used to indicate errors and they still prevent the user from performing some actions that they
might otherwise wish to perform. The dialogue box might be asking the user to
fill in some information to specify parameters for a command. If the user does not
know what to provide, then they are stuck. A lot of the time, the user can find out
the information by browsing through some other part of the system, but in order
to do that they must exit the dialogue box (and forfeit any of the settings that they
might have already entered), find out the missing information and begin again.
This kind of preemption is not desirable. It is probably the kind of preemption this
guideline is intending to prevent, but it doesnt always get applied.

Exercise 5.4
Find a book on guidelines. List the guidelines that are provided and
classify them in terms of the activity in the software life cycle to which
they would most likely apply.

June 4, 1993 17:34

exercises

Sheet number 33 Page number 31

The Design Process

Answer
We use as a source of guidelines Mayhews book Principles and Guidelines in Software
User Interface Design. In general, all guidelines offer constraints on the design
activity and so should be known during requirements phase. In the following list,
we will concentrate on what other stages (architectural design, detailed design,
coding and unit testing, integration and testing) will be most affected by the
guideline. The numbers in parentheses indicate the page reference for the given
guideline.

Architectural design
Present functionality through a familiar metaphor. (97)
Provide similar execution style of analogous operations in different applications. (97)
Organize the functionality of a system to support common user tasks.
(442)
Make invisible parts and processes visible to the user. (95)

Detailed design
Consistent dialogue style for different functions. (97)
Match menu structure to task structure. (144)
Create logical, distinctive and mutually exclusive semantic categories
with clear meanings. (150)
Design and organize a fill-in form to support the task. (184)
Consider voice synthesis as an output device when the users eyes are
busy, when mobility is required, or when the user has no access to a
workstation or screen. (427)

Coding and unit testing

On full-screen text menus, present menu choice lists vertically. (148)
In a fill-in form, use white space to create a balance and symmetry and
lead the eye in the appropriate direction. (186)
Avoid frequent use of shift or control keys. (256)
Place high-use function keys within easy reach of the home row on the
keyboard. (281)

Integration and testing

Allow full command names and emphasize them in training, even if
abbreviations are allowed. (261)

31

June 4, 1993 17:34

32

exercises

Sheet number 34 Page number 32

The Design Process

Exercise 5.5
What is the distinction between a process-oriented and a structure-oriented
design rationale technique? Would you classify psychological design rationale as process- or structure-oriented? Why?

Answer
The distinction between process- and structure-oriented design rationale resides
in what information the design rationale attempts to capture. Process-oriented
design rationale is interested in recording an historically accurate description of
a design team making some decision on a particular issue for the design. In this
sense, process-oriented design rationale becomes an activity concurrent with the
rest of the design process. Structure-oriented design rationale is less interested in
preserving the historical evolution of the design. Rather, it is more interested in
providing the conclusions of the design activity, so it can be done in a post hoc and
reflective manner after the fact.
The purpose of psychological design rationale is to support the task-artifact
cycle. Here, the tasks that the user community performs are changed by the systems
on which they perform the tasks. A psychological design rational proceeds by
having the designers of the system record what they believe the tasks are that the
system should support and then building the system to support the tasks. The
designers suggest scenarios for the tasks which will be used to observe new users
of the system. Observations of the users provide the information needed for the
actual design rationale of that version of the system. The consequences of the
designs assumptions about the important tasks are then gauged against the actual
use in an attempt to justify the design or suggest improvements.
Psychological design rationale is mainly a process-oriented approach. The
activity of a claims analysis is precisely about capturing what the designers assumed about the system at one point in time and how those assumptions compared to actual use. Therefore, the history of the psychological design rationale is
important. The discipline involved in performing a psychological design rationale
requires designers to perform the claims analysis during the actual design activity,
and not as post hoc reconstruction.

June 4, 1993 17:34

exercises

Sheet number 35 Page number 33

Chapter 6

Models of the User in Design

Exercise 6.1
Create a GOMS description of the task of photocopying a paper from a
journal. Discuss the issue of closure in terms of your GOMS description.

Answer
One possible GOMS description of the goal hierarchy for this task is given below.
Answers will vary depending on assumptions about the photocopier used as the
model for the exercise. In this example, we will assume that the article is to be
copied one page at a time and that a cover over the imaging surface of the copier
has to be in place before the actual copy can be made.
GOAL:
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

PHOTOCOPY-PAPER
GOAL: LOCATE-ARTICLE
GOAL: PHOTOCOPY-PAGE repeat until no more pages
.
GOAL: ORIENT-PAGE
.
.
OPEN-COVER
.
.
SELECT-PAGE
.
.
POSITION-PAGE
.
.
CLOSE-COVER
.
GOAL: VERIFY-COPY
.
.
LOCATE-OUT-TRAY
.
.
EXAMINE-COPY
GOAL: COLLECT-COPY
.
LOCATE-OUT-TRAY
.
REMOVE-COPY (outer goal satisfied!)
GOAL: RETRIEVE-JOURNAL
.
OPEN-COVER
.
REMOVE-JOURNAL
.
CLOSE-COVER
33

June 4, 1993 17:34

34

exercises

Sheet number 36 Page number 34

Models of the User in Design

The closure problem which appears in this example occurs when the copy of
the article is removed from the photocopier out tray, satisfying the overall goal for
the task. In the above description, however, the original journal article is still on the
imaging surface of the photocopier, and the cover is closed. The user could easily
forget to remove the journal. How could the photocopying procedure be revised
to eliminate this problem? One answer is to force the goal RETRIEVE-JOURNAL
to be satisfied before COLLECT-COPY.

Exercise 6.2
Recall the CCT description of the rule INSERT-SPACE-2 discussed in
Section 6.3.2:
(INSERT-SPACE-2
IF (AND (TEST-GOAL insert space)
(TEST-CURSOR %LINE %COL) )
THEN ( (DO-KEYSTROKE I)
(DO-KEYSTROKE SPACE)
(DO-KEYSTROKE ESC)
(DELETE-GOAL insert space) ))
As we discussed, this is already proceduralized, that is, the rule is an
expert rule. Write new novice rules where the three keystrokes are not
proceduralized. That is, you should have separate rules for each keystroke
and suitable goals (such as GET-INTO-INSERT-MODE) to fire them.

Answer
(INSERT-SPACE-BEGIN-SET-MODE
IF (AND (TEST-GOAL insert space)
(TEST-CURSOR %LINE %COL)
(TEST-NOTE in command mode ))
THEN ( (ADD-GOAL get into insert mode)))
(INSERT-SPACE-END-SET-MODE
IF (AND (TEST-GOAL insert space)
(TEST-GOAL get into insert mode))
THEN ( (DO-KEYSTROKE I)
(ADD-NOTE in insert mode)
(DELETE-GOAL get into insert mode)))
(INSERT-SPACE-DOIT
IF (AND (TEST-GOAL insert space)

June 4, 1993 17:34

exercises

Sheet number 37 Page number 35

Models of the User in Design

THEN (

(TEST-NOTE in insert mode)

(TEST-CURSOR %LINE %COL))
(DO-KEYSTROKE SPACE)
(ADD-GOAL get into command mode)))

(INSERT-SPACE-CLEAN-UP
IF (AND (TEST-GOAL insert space)
(TEST-NOTE in insert mode)
(TEST-GOAL get into command mode))
THEN ( (DO-KEYSTROKE ESC)
(DELETE-GOAL get into command mode)
(DELETE-GOAL insert space)
(DELETE-NOTE in insert mode)
(ADD-NOTE in command mode)))

Exercise 6.3

Do a keystroke level analysis for opening up an application in a visual

desktop interface using a mouse as the pointing device, comparing at least
two different methods for performing the task. Repeat the exercise using
a trackball. Discuss how the analysis would differ for various positions of
the trackball relative to the keyboard and for other pointing devices.

Answer
We provide a keystroke level analysis for three different methods for launching an
application on a visual desktop. These methods are analyzed for a conventional
one-button mouse, a trackball mounted away from the keyboard and one mounted
close to the keyboard. The main distinction between the two trackballs is that the
second one does not require an explicit repositioning of the hands, that is, there
is no time required for homing the hands between the pointing device and the
keyboard.

Method 1: Double clicking on application icon.

Steps
1. move hand to mouse
2. mouse to icon
3. double click
4. return to keyboard
Total times

Operator
H[mouse]
P[to icon]
2B[click]
H[kbd]

Mouse
0.400
0.664
0.400
0.400
1.864

Trackball1
0.400
1.113
0.400
0.400
2.313

Trackball2
0.000
1.113
0.400
0.000
1.513

35

June 4, 1993 17:34

36

exercises

Sheet number 38 Page number 36

Models of the User in Design

Method 2: Using an accelerator key
Steps
Operator
1. move hand to mouse H[mouse]
2. mouse to icon
P[to icon]
3. click to select
B[click]
4. pause
M
5. return to keyboard
H[kbd]
6. press accelerator
K
Total times

Mouse
0.400
0.664
0.200
1.350
0.400
0.200
3.214

Trackball1
0.400
1.113
0.200
1.350
0.400
0.200
3.663

Trackball2
0.000
1.113
0.200
1.350
0.000
0.200
2.763

Method 3: Using a menu

Steps
1. move hand to mouse
2. mouse to icon
3. click to select
4. pause
5. mouse to File menu
6. pop-up menu
7. drag to open
8. release mouse
9. return to keyboard
Total times

Mouse
0.400
0.664
0.200
1.350
0.664
0.100
0.713
0.100
0.400
4.591

Trackball1
0.400
1.113
0.200
1.350
1.113
0.100
1.248
0.100
0.400
6.024

Trackball2
0.000
1.113
0.200
1.350
1.113
0.100
1.248
0.100
0.000
5.224

Operator
H[mouse]
P[to icon]
B[click]
M
P
B[down]
Pdrag
B[up]
H[kbd]

Exercise 6.4
One of the assumptions underlying the programmable user model approach is that it is possible to provide an algorithm to describe the users
behaviour in interacting with a system. Taking this position to the extreme,
choose some common task with a familiar interactive system (e.g. creating
a column of numbers in a spreadsheet and calculating their sum, or any
other task you can think of) and describe the algorithm needed by the user
to accomplish this task. Write the description in pseudocode. Does this
exercise suggest any improvements in the system?

Answer
This is a pretty open-ended exercise, so no model answer is provided.

June 4, 1993 17:34

exercises

Sheet number 39 Page number 37

Chapter 7

Task analysis

The exercises for this chapter, like the chapter itself, concentrate on real world
rather than computer examples. This is largely because task analysis of current
computer systems depends on the particular systems available. However, the form
of a general student exercise could be:
Observe use of the computer system. Perform a task analysis (of some
kind). To what extent do you think the way people perform these tasks is
determined by the system and to what extent by the fundamental aspects
of the task? Does the systems menu layout etc. support the tasks it is used
for, in particular, are frequent task sequences easy to perform? Suggest
potential improvements, both incremental changes and radical redesigns
of the system.
The details of the question can be of course varied depending on the system and
the students can be directed to particular sub-systems where you have observed
problems.

Exercise 7.1

The following is a list of objects found in one of the authors kitchens.

teapot, mug, soup bowl, plate, spoon, table knife, cooks knife,
fork, saucepan, frying pan, kettle, casserole, fish slice, tin opener,
baking tray, scales, miking bowl, glasses, jugs, corkscrew, rolling
pin, ladle, egg cup, chopping board
Produce a taxonomy using the TDH notation of these objects. Does it obey
the TAKD uniqueness rule? Compare your answer with someone elses.
(Note, the authors had great difficulty with items like the corkscrew, which
did not fit easily into any generic category perhaps you did better.)
37

June 4, 1993 17:34

38

exercises

Sheet number 40 Page number 38

Task analysis
kitchen object XOR
preparation XOR
pre-preparation XOR
opening
tin opener, cork screw
measuring
scales, (measuring)jug
proper preparation XOR
active
rolling pin, cooks knife, (cooks) spoon
passive
mixing bowl, chopping board
cooking XOR
passive
teapot
active XOR
external power
saucepan, frying pan, casserole, baking tray
internal power
(electric) kettle
serving XOR
serving
fish slice, (serving) jug, ladle
eating XOR
active
spoon, fork, knife
passive XOR
food
egg cup, soup bowl, plate
drink
mug, glass

j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j

j
j
j
j
j
j
j
j
j
j
j
j

j
j
j
j
j
j
j
j
j
j
j
j

j
j
j

Figure 7.x1 TDH taxonomy produced by first subject

Answer
As the authors had already produced a partial taxonomy, we interviewed two
domain experts (cooks). They were asked to describe how they would group and
classify the kitchen items. they were explicitly told (and reminded) that they could
have multiple classifications and put the same item into several categories. The
authors then cast their answers into TDH notation.
One of the subjects was a doctor and used to medical taxonomies of disease.
Despite stressing the looseness of the classifications, he insisted on a complete
taxonomic tree (Figure 7.x1). As you see all his branches are XOR branches. On
discovering that jug had to fit in two places in his taxonomy, he split it into

June 4, 1993 17:34

exercises

Sheet number 41 Page number 39

Task analysis
kitchen object OR
{
things for making tea
{
teapot, mug, kettle, spoon
things for eating meals
{
{
soup bowl, plate, glasses, egg cup
cutlery for meals
{
{
spoon, table knife, fork
cutlery for cooking
{
{
spoon, fork, fish slice, tin opener, table knife,
{
corkscrew, rolling pin, ladle
things for making meals
{
{
saucepan, frying pan, casserole, baking tray, scales,
{
mixing bowl, jugs, chopping board
things for serving meals
{
jugs, casserole, fish slice, corkscrew, ladle, spoon
Figure 7.x2 Initial version of TDH taxonomy produced by second subject

serving jug and measuring jug. This emphasises the need for the task analyst
rather than the domain expert to actually draw up the taxonomy!
As it is a true tree it clearly does not satisfy the uniqueness rule, but the only
way it could is to invent spurious new categories. One could under opening add
categories for bottles (containing corkscrew) and tins (continuing tin opener), but
this would not improve clarity.
If the first subject was a stickler for precision the second subject, preferred
broad categories. Figure 7.x2 shows her initial classification. We wanted to put
some additional structure on this and so, after some discussion, the subject agreed
that her basic distinctions were those of function (making meals etc.) and between cutlery and non-cutlery. Using these to form an AND branch, we obtained
Figure 7.x3.
This taxonomy does not obey the uniqueness rule either, for example, fish slice
and ladle always appear together. In terms of KRG they are both:
kitchen object/material(cutlery)/
function{cooking meals,serving meals}/
The TAKD purist might demand extra categories to fulfill the uniqueness rule.
However, the authors would recommend that students simply be taught to recognise the rule and use it as a heuristic.
It is interesting that both domain experts focussed on the functional view of
the items, just as the authors did in the book. This suggests that it is indeed a
generic way of classifying kitchen objects and would thus be a good candidate for
classification in a catalogue or menu system. The second subject also noted that
her original breakdown was inspired, not so much by the function per se, but by

39

June 4, 1993 17:34

40

exercises

Sheet number 42 Page number 40

Task analysis
kitchen object AND
/
material XOR
cutlery
/
spoon, table knife, fork, fish slice, tin opener
/
/
corkscrew, rolling pin, ladle
non-cutlery
/
/
teapot, mug, kettle,
/
soup bowl, plate, glasses, egg cup,
/
saucepan, frying pan, casserole, baking tray, scales,
/
mixing bowl, jugs, chopping board
function OR
/
making tea
{
{
teapot, mug, kettle, spoon
cooking meals
{
{
spoon, fork, fish slice, tin opener, table knife,
{
corkscrew, rolling pin, ladle,
{
saucepan, frying pan, casserole, baking tray, scales,
{
mixing bowl, jugs, chopping board
serving meals
{
{
jugs, casserole, fish slice, corkscrew, ladle, spoon
eating meals
{
soup bowl, plate, glasses, egg cup,
spoon, table knife, fork

j
j
j
j

Figure 7.x3 Refined version of TDH taxonomy produced by second subject

where they were stored in her kitchen itself determined largely by function. This
is perhaps the physical equivalent of a menu system!

Exercise 7.2
Complete the tea-making manual in Figure 7.7. Do you think it would be
useful? Think of situations where such a manual would be helpful and
where a more conceptual manual would be better.

Answer
Although a manual for the tea making might be regarded as a little extreme, such
manuals are useful in several situations. You could pose this exercise, together
with the initial task analysis, for different domains where more of the following
situations are pertinent.
The first situation where a procedural manual is useful is for the absolute
novice who has no idea of the conceptual background. This might be a first time

June 4, 1993 17:34

exercises

Sheet number 43 Page number 41

Task analysis
user or may be for an infrequently performed activity. A good example of the latter
is the installation of computer equipment, which most users perform only once
every couple of years. Similarly, recipe books are laid out in a highly procedural
fashion, although unfortunately not always clearly:
Beat the egg whites until they froth, then put them into a ramekin. While
beating the egg whites, slowly add the white wine.
The second situation is where there is some sort of safety critical aspect and
errors, however well thought out, can be disastrous. Often in such a situation, the
additional stress can cloud judgement and make it far safer to stick to a predetermined drill. One example, of this are emergency procedures in large chemical
or nuclear installations when an emergency arises the operators are expected
to stick closely to the set procedures. The accident at Chernobyl came about in
part because the operators felt that they knew enough to override the rule book.
Reading a manual in such circumstances may be too time consuming, but an HTA
can be used to train the operator to respond automatically. The use of HTA for
military training is largely in this vein.
Thirdly, the situation may not be safety critical, but may be time critical. Much
analysis may have gone into discovering the most efficient manner to perform a
task and that way is then taught, by rote, to the operators. Although this form of
time and motion approach is less likely to be useful in an information intensive
job than in a factory (if there!), there are jobs, such as telephony, where it is still
important.
Finally, the user may not have sufficient knowledge to understand why a
process works, but can follow a set of instructions. This may either relate to the
complexity of the task or the skill of the operator. If one were teaching kitchen
craft to the mentally handicapped, then just such a procedural description of tea
making would be required.
The problem with such procedural manuals is that they give the operator no
real feeling as to why the tasks are performed in the way they are. Whether such a
manual is preferred by a novice user depends very much on the users personality.
Some people prefer to have a set of instructions to get them started, whereas
others find it very difficult to use something without some sort of conceptual
understanding.
The procedural manual really becomes unstuck when the set of tasks considered are not complete. When faced with a radically new task the user must
understand enough of the domain to perform it ad hoc or to modify an existing procedure. One frequent cause of entirely new situations is unforeseen breakdowns
of equipment. For example, if the kettle was broken, one could then abstract that
the real reason for boiling the kettle was to heat water and that this could be performed by heating a bowl of water in the microwave oven. Such a modification of
the procedure is not even suggested by the procedural manual.

41

June 4, 1993 17:34

42

exercises

Sheet number 44 Page number 42

Task analysis
0.
clean the
house
plan 0.
1235
when the dust bag gets full 4
1.
get vacuum
cleaner out

2.
fix appropriate
attachment

3.

4.

clean rooms

5.
empty the
dust bag

put away
everything

plan 3.
do 3.1 every day
3.2 once a week
when visitors are due 3.3
3.1.

3.2.
clean
the hall

clean
living rooms

3.3.
clean
bedrooms

Figure 7.x4 HTA diagram for vacuum cleaning a house

Exercise 7.3
Figure 7.1 shows a textual representation of an HTA description of vacuum
cleaning. Present the same information in a diagrammatic form.

Answer
This is given in Figure 7.x4. In fact, this uses the revised version of plan 3, but the
choice of plan makes no difference to the structure of the diagram.

Exercise 7.4
(Converse to above) Figure 7.6 uses an HTA diagram to show the actions
in which a tractor is involved; show the same information textually.

June 4, 1993 17:34

exercises

Sheet number 45 Page number 43

Task analysis

Answer
0.

life cycle of tractor

1.
maintenance
2.
cultivation
2.1. connect implement
2.1.1 fix harrow
2.1.2 fix plough
2.2. drive to field
2.3. cultivate field
2.4. drive to shed
2.5. put away

Plan 0:

as required 2
when tractor breaks down 1

Plan 2:

2.1 2.2 2.3 2.4 2.5

Plan 2.1: one of 2.1.1 or 2.1.2 depending on job

Exercise 7.5
Observe an office, note the actions performed and the objects used depending on the available equipment, use different recording techniques as
described in Chapter 11. Then use the different task analysis techniques
to structure your findings. (Note, this could be a group project.)

Answer
The easiest starting point for the students is simply to go around the office noting
down what is there:
typewriter, corrector fluid, desk diary, pen, pencil, scissors, envelopes,
paper clips, typing paper, post-it notes, telephone, telephone directory (internal and external), filing cabinet containing folders, clock, wall calendar.
This list can then be used to begin to build both a knowledge based or an
entity-relation description. However, the later will also require at least a list of
actors. In a university office this might include the following:
secretary, lecturer, student (undergraduate), research student, research
staff, administrator

43

June 4, 1993 17:34

44

exercises

Sheet number 46 Page number 44

Task analysis
However, the roles that they take may not be simple. For example, we may
find that a lecturer comes into the office to use the typewriter. That is the lecturer
acts in the role of typist.
Neither of these descriptions can be complete, nor can an HTA begin, without
a list of activities. This can be obtained in two main ways. First, the students
can simply make an unstructured list of all the activities they see and then add
structure to it. Alternatively, they can follow specific tasks noting what is done
in what order. In the later case, they should be encouraged to write the list of
activities in a purely sequential manner they are observing. Only later will they
build upon this an hierarchical interpretation.
It may obviously cause severe inconvenience if all the members of a class were
to interview the office staff. However, to gain first hand interview experience, some
domain expert can be invited into class to talk about their work and be questioned
about it. Alternatively, students could make their own notes from a preprepared
video taped interview.
If a question of this sort is used as an assessment, then we would suggest that
the students hand in not just the completed task analysis, but intermediate notes
and representations. The most important thing in determining the effectiveness of
their analysis is the care with which they carried out the original observation and
subsequent working.

Exercise 7.6
Consider the activity of making a telephone call. Record the actions in
an HTA diagram or textually. Start off simply, assuming you know the
number to dial, but then add more complicated situations: finding the
number in an address book, or what to do when the number is engaged.

Answer
As with most of these exercises, this is an open ended question. Here is a simple
version with some expansion, but one can look at alternatives such as public payphones.
0.

make phone call

1.
pick up receiver
2.
dial number
3.
wait for reply
4.
talk
5.
replace receiver

June 4, 1993 17:34

exercises

Sheet number 47 Page number 45

Task analysis
Plan 0:

123
when answered 4
when finished 5

We now add looking up the number. The form this takes depends on whether
we find the number in an address book or a telephone directory. If both fail, say if
the call is long distance to someone not in a local directory, the telephone operator
must be consulted. If students have some form of online telephone list, possibly
in a palmtop computer, they could included the analysis of the lookup procedure.
The next exercise, dictionary lookup, is very similar to looking up a number in a
directory, and so further analysis of these branches is not included here. Note that
contacting the telephone operator involves making a phone call, but the steps for
this are not repeated in full!
0.

make phone call

1.
find number
1.1. look in address book
1.2. look in phone directory
1.3. ask operator
1.3.1 pickup receiver

2.
actually call
2.1. pick up receiver
2.2. dial number
2.3. wait for reply
2.4. talk
2.5. replace receiver

Plan 0:

if number unknown 1
when number found 2

Plan 1:

if phoning friend 1.1

if local call 1.2
if 1.1 or 1.2 fail 1.3

Plan 2:

2.1 2.2 2.3

when answered 2.4
when finished 2.5

Finally, we add the case when the phone is engaged. The simplest way to do this
is simply to change Plan 2.
Plan 2:

2.1 2.2 2.3

if answered 2.4 then when finished 2.5
if engaged 2.5

45

June 4, 1993 17:34

46

exercises

Sheet number 48 Page number 46

Task analysis
However, looking at the second line it might suggest that we modify 2.4 to have
two parts:
0.

make phone call

2.
actually call

2.4. successful call

2.5.1. talk
2.5.2. replace receiver
2.5. failed call
2.5.1. replace receiver

Plan 2:

2.1 2.2 2.3

if answered 2.4
if engaged 2.5

Plan 2.4: 2.4 then when finished 2.5

Exercise 7.7
The act of looking up a person in an address book or telephone directory is
itself quite complicated. Get several friends to look up words in a dictionary. Observe closely their methods. You will probably have to develop
shorthand notations to keep track of what pages they visit. Compare the
strategies used by the different people. If they differ, try to abstract out
the common parts of the task and the variable parts. Those with sufficient
computing background can attempt to classify their methods in relation
to known search algorithms: binary chop, linear search, etc.

Answer
In order to see the range of possible search strategies,the authors asked two subjects
to look up words in a dictionary. One was a literate, but young, child and the other
an erudite English graduate.
To obtain any sort of useful record, the subjects had to be asked to work slowly
and even then keeping track of the search was difficult. We noted down the page
numbers they paused at, but often searches included thumbing through a sequence
of pages, at which point some form of specific notation would be useful. As an
alternative to noting down the page numbers, noting the words at the top of the
page could be used.

June 4, 1993 17:34

exercises

Sheet number 49 Page number 47

Task analysis
????
283 dog

note illegible, intended to convey that she turned

the dictionary on its side and then opened it
said that first cut was based on the position
of d in the alphabet

flick backward
overshoot to C
forward one page to beginning of D
quick skim over page
places finger on word

daltonism on this page

Table 7.x1 Adult looking up the word daltonism

right of centre
comment by subject on position of p in dictionary
729 O
gets to page in the middle of O section
turn over forward
741
page begins with para words
? C or K
subject says unsure whether spelt with c or k
says shell look for c first
scan with finger
sequential
she explains that the presence of the para words
made the task confusing

Table 7.x2 Adult looking up the word parclose

The English graduate was also asked to talk aloud through the exercise. We
discuss in Chapter 11 the possible effects this might have on the users performance,
but it is extremely difficult to keep track of the readers focus on the page otherwise.
The notes below are tidied up considerably from our hand written scrawl! The
notes in italics were added after. Unfortunately, in typesetting the notes, some of
the spatial layout is lost. They do, however, retain the scrappiness of the original
observations. Use of video can of course allow one to replay tasks at a slower rate
and thus improve the detail of note taking.
The English graduate was asked to find unfamiliar words: daltonism (Figure 7.x1) and parclose (Figure 7.x2). The child was asked to lookup simpler
words: dig (Figure 7.x3) and orange (Figure 7.x4). The transcripts do not include timings the whole process took no more than 30 seconds. To give an idea
of relative timings: whereas the English graduate homed quickly onto the correct
page, this took the child somewhat longer, but, once they were on the right page,
both were so quick that it is hard to give judgement on relative speed.
From this evidence, about the best general description we could have is:

47

June 4, 1993 17:34

48

exercises

Sheet number 50 Page number 48

Task analysis

525
429
297
266
scan down and up

near middle of book

these first pages were quite rapid
page headed dualism
correct page luck!
sequential search item by item

Table 7.x3 Child looking up the word dig

236
266
flick forward
800
flick back
700, 690, 699
sequential pages
703
flick forward page by page
717
sequential search through page

last page of c
too fast to note pages
overshoot

Table 7.x4 Child looking up the word orange

start of O

right page

June 4, 1993 17:34

exercises

Sheet number 51 Page number 49

Task analysis
0.

look up word in dictionary

1.
find right page
2.
find word on page

Plan 0:

12

The performance of subtask 1 was far more efficient for the adult as she used a
heuristic to find the approximates position based on her knowledge of the alphabet.
The child knew the relative position of words in the alphabet, but not enough to be
able to say p is just after the middle as the adult did. The differences in subtask 2
was less marked, but mainly because it was too rapid to annotate. There was some
evidence that the adult used a more direct method than the child who simply (but
quickly) scanned the page entry by entry.
If individual students observe more subjects, or as they come together to make
notes, they may be able to abstract more specific, but still general search strategies.
For example, they may see something akin to binary chop. They should not look
for a single strategy used by all people, but instead look for a small set of strategies,
and then look for ways of telling how people choose one strategy or another.
Really, two observations, as described above, is only sufficient for a pilot study
in order to sharpen up thinking. Looking at the notes may suggest shorthands,
for instance, arrows to represent different kinds of scanning. It also prompts one
to ask questions. What happens if when they scan the page at step 2, the subjects
discover that they are still on the wrong page? Presumably this leads to a new
find the right page step. However, this subtask, essentially a fine correction, will
be very different from the original search.
0.

look up word in dictionary

1.
find right page
2.
find word on page
3.
adjust page

Plan 0:

12
if still wrong page 3 2

However, unlike the first HTA, this one is not based on observation, but on the
authors imagination. Given such refined expectations, the students ought to return
to observation in order to explore further. Perhaps, they could deliberately choose
words close to page boundaries in order to confuse the subjects. Alternatively,
they could ask their subjects what would you do if youd got the wrong page?.
You should emphasise to the students the difference between a psychological
experiment in problem solving and a task analysis observation. For the former, we
would be simply trying to verify our guess/theory. For the latter, we might have
an idea, but our observations should be open ended as we want to find out what
really happens, whether or not it agrees with our theories.
Although the question is clearly oriented towards procedural analysis, One
could also do some form of knowledge based analysis. Clearly concepts such as

49

June 4, 1993 17:34

50

exercises

Sheet number 52 Page number 50

Task analysis
position in the alphabet and word at the top of the page are important. Indeed,
if we were designing an electronic dictionary such concepts may be more useful
than emulating paper.
Indeed, one could extend this question into a design exercise how to design
the electronic dictionary. Assuming that we are not interested in totally automatic
lookup, how would we design the interaction? Whereas in a book, ones normal
action is to turn over one page backwards or forwards, in dictionary lookup one
jumps into the middle, turns over several pages at a time and flicks through pages
at high speed. This can emphasise that a question like how do we design an
electronic book? cannot be answered until we know for what purposes that book
is to be used.

June 4, 1993 17:34

exercises

Sheet number 53 Page number 51

Chapter 8

Dialogue Notations and Design

Exercise 8.1
Complete the drawing tool STN in Figures 8.1 and 8.3 by writing dialogue
descriptions for the text and paint sub-menus. For the text sub-menu
assume that there are three options: centred, left and right justified. The
text is entered by clicking at a location in the drawing surface and then
typing. You may initially assume that typing a line of text can be regarded
as a single user action. But later try regarding each character typed as
an action. The paint sub-menu has two options: a pencil for free hand
drawing and a paint pot for flood filling. The former is performed by
holding the mouse button down whilst moving the mouse about to draw
the line. The paint pot is activated by simply clicking the mouse over the
area to be filled.

Answer
The STNs for the main menu and graphics subsystem are in the text. We only need
to do the text-submenu and the paint-submenu. For each the overall structure
is similar to the graphics submenu, that is, the user selects an option, the STN
branches and then there is a description of each options dialogue.
First do the text sub-menu considering typing a line of text as a single user
action. This is shown in Figure 8.x1. To consider individual letters being typed,
we simply add a loop terminated by the user typing the enter key (Figure 8.x2).
Finally, we do the paint sub-menu in Figure 8.x3.
Note that the fundamental actions for the pencil drawing are pressing and
releasing the mouse button, rather than clicking it. Remember the more complex
behaviour of the watch when we considered button press and release separately. It
is worth discussing the expected behaviour of the package on other options during
mouse depression. For example, when drawing lines, which position is registered
the one when the button goes down, or the one when it is released. Encourage
the students to try it out on different drawing packages. There is a similar issue
51

June 4, 1993 17:34

52

exercises

Sheet number 54 Page number 52

Dialogue Notations and Design

select
left justified
highlight
left justified
Start

Text
Menu

Left 1

click on location
show text cursor

Left 2

type line of text

remove text cursor

Finish

...
...

Figure 8.x1 STN of text submenu

from
Left 1

Left 2

type ENTER
remove text cursor

Finish

type character
display character

Figure 8.x2 Character by character input

select paint pot

highlight paint pot

Start

Pot 1

click inside shape

fill shape with colour

Finish

Paint
Menu
depress
select pencil
mouse button
Pencil 1
Pencil 2
highlight pencil
change mouse
cursor to pencil
move mouse
colour pixels

Figure 8.x3 Paint submenu

release
mouse button
restore
mouse cursor

Finish

June 4, 1993 17:34

exercises

Sheet number 55 Page number 53

Dialogue Notations and Design

with menu selection and screen buttons. The scenario described in Section 9.4.5 is
an example where the important location is where the mouse button is released.
Again get the students to look at different GUIs and their detailed interaction
properties. You may be surprised at the differences!

Exercise 8.2
Repeat the above exercise using different notations, grammars, production
rules, JSD or CSP. You will need to specify the whole system from the main
menu to the individual sub-menu selections such as circle drawing. Note
the problems you have with each notation.

Answer
Below are partial descriptions in BNF and production rules. As a class exercise,
you could allocate different notations to different groups of students and then
compare the answers in different notations.
BNF description
The BNF description is quite straightforward (if a little tedious). We give the
description for the main menu, part of the graphics submenu and the text submenu. The line drawing is already described in section 6.4.1 as is the definition of
position-mouse.
drawing-tool
main-menu

::=
::=

main-menu + drawing-tool
graphics-submenu
| text-submenu
| paint-submenu

graphics-submenu ::= draw-circle | draw-line

draw-circle
::= select-circle + choose-centre + choose-edge
select-circle ::= position-mouse + CLICK-MOUSE
choose-centre ::= position-mouse + CLICK-MOUSE
choose-edge
::= position-mouse + CLICK-MOUSE
text-submenu
::= select-left
+ choose-location + type-text
| select-centre + choose-location + type-text
| select-right + choose-location + type-text
select-left
::= position-mouse + CLICK-MOUSE
select-centre
::= position-mouse + CLICK-MOUSE
select-right
::= position-mouse + CLICK-MOUSE

53

June 4, 1993 17:34

54

exercises

Sheet number 56 Page number 54

Dialogue Notations and Design

choose-location ::=
type-text
::=

position-mouse + CLICK-MOUSE
ENTER-KEY
| PRINTABLE-CHAR + type-text

Notice the different styles possible. In the graphics submenu, the option
selection and the action selections are each packaged up into a named sub-dialogue
(draw-line and draw-circle). However, in the text-submenu the top-level
description of each option is included within the definition of text-submenu.
Emphasise to the students that, with all notations, there is a wide degree of stylistic
choice. Suitably named sub-dialogues can make the entire dialogue far easier to
read, but, on the other hand, too many levels of abstraction can be confusing.
Also notice that the BNF description does not distinguish between selecting
a menu option or choosing a point on the drawing surface. Both are described
as position-mouse followed by CLICK-MOUSE. At the level of user movements
and keystrokes, this is perfectly accurate, but loses the clearly perceived difference.
This might be a good time to introduce your students to the UAN notation (Hartson
et al. 1990)

Production rules
As the drawing tool is largely sequential, we would expect to find the production
rule description rather cumbersome. The main problem is keeping track of how
deep you are in the stack of menus submenus and options. Watch out for solutions
where the higher level options are still active, when the lower level dialogues are
operating. For example, the top level description might read:
Sel-graphics
Sel-text
Sel-paint

!
!
!

<pop-up
<pop-up
<pop-up

>

graphics menu
text menu
paint menu

>
>

This could exhibit strange behaviour. Imagine that the user is in the middle of
drawing a line (using the first description in section 8.4.2). Then after selecting the
first point (at which point the system event rest-line is active), the user goes
back to the top level menu and selects text. At this point the text pup-up menu
appears, and the user is able to start putting text annotations on the drawing area:
Sel-left-just
C-point left-start
Type-line left-type

!
!
!

left-start
left-type typing cursor on
put text left justified
typing cursor off

<

<

<

>

>

>

The user selects left justified and goes to the drawing surface and clicks at a
point. However, at this point there are two rules which can fire.

C-point left-start from the text rules (above), setting the location for

typing.

June 4, 1993 17:34

exercises

Sheet number 57 Page number 55

Dialogue Notations and Design

C-point rest-line from the drawing rules in Section 8.4.2,putting another

point on the line.

Whereas it is certainly the first of these that the user expects, the second is
possible because the rest-line event is still active. Also looking back to the
definitions in section 8.4.2, we see that rubber banding would still be on during
the entire text menu selection process.
There are various fixes for this that you can discuss with your students. One
way is to use clean up rules as used in CCT in section 6.3.2. For example, we
would modify the top level description so that it is not always active:

!
!
!

Sel-graphics main-active
graphics-active pop-up graphics menu
Sel-text main-active
text-active pop-up text menu
Sel-paint main-active
paint-active pop-up paint menu

<

<

<

>

>

>

The submenu options are then altered to update these activity events (the
semantic actions have been omitted for clarity):
Sel-line graphics-active
C-point start-line
C-point rest-line
D-point rest-line

!
!
!
!

start-line
rest-line
rest-line
main-active

This now ensures that the user finished low-level dialogues before selecting
new options at the top level.

Exercise 8.3
Develop the JSD diagram in Figure 8.12, expanding the various nodes until
you get to basic operations such as prompt "login" or user types in
password. Expand the delete employee node using the dialogue style as
described in Figure 8.11, and use your imagination for the rest.

Answer
This is a fairly simple exercise, and some of the nodes are expanded here. In
Figure 8.x4, the login sub-dialogue is expanded, but it assumes that the user types
the correct password during the login process. To change it to accommodate
incorrect passwords would be inordinately complex for two reasons. Firstly, the
JSD diagram does not distinguish between user choice (whether to add or delete
a record) and system choice. This is evident in both the delete and change subdialogues (Figures 8.x6 and 8.x5). In each of these the actual updating of the file

55

June 4, 1993 17:34

56

exercises

Sheet number 58 Page number 56

Dialogue Notations and Design

login

get
user name

prompt
login:

get password

read
user name

prompt
passwd:

read
password
(no echo)

verify
password

Figure 8.x4 JSD diagram for logging in to personnel system

is put in an optional box as the user may have answered N (no) when asked to
confirm the update. Secondly, making the dialogue dependent on the password
would mean that the rest of the dialogue would sit under an option box with
another box to abort. This was acceptable in the delete and change sub-dialogues,
but if this were applied to the login sequence, the resulting dialogue would become
degenerate and hide the normal hierarchical structure. This problem of whether
or not to include exceptional cases and error behaviour is extremely complex. In
early stages of design it is the normal cases which one wants to consider. Later on
one might refine the dialogue descriptions, or alternatively annotate the normal
description.

Exercise 8.4

Using CSP, construct a dialogue for one user with an application in one
window of a multi-window system. Using this first general CSP description, provide a CSP description for a multi-window interaction using the
parallel operator (||). How does this approach handle interference and
information sharing between windows?

June 4, 1993 17:34

exercises

Sheet number 59 Page number 57

Dialogue Notations and Design

57

change
employee
record

get
employee
number

read and
display
record

user
edits
record

update
record
in file

confirm
changes

...

...
select
field

edit
field
contents

Figure 8.x5 JSD diagram for updating a record

delete
employee
record

get
employee
number

read and
display
record

confirm
delete

prompt
delete ? (Y/N)

incorrect
answer

delete
record
from file

...
*

user doesnt
answer
Y or N

user
answers
Y or N

error
message

Figure 8.x6 JSD diagram for record deletion sub-dialogue

June 4, 1993 17:34

58

exercises

Sheet number 60 Page number 58

Dialogue Notations and Design

Answer
The intention of this exercise is not so much to produce a clean description, but
to show how complex the business of talking about multi-window applications is.
Most interface designers and programmers dont have to worry about designing
the windowing system itself. However, it does help students to appreciate why
areas like cut-paste buffers can behave oddly.
The applications can be anything. In fact, you may like to give the students
some simple applications (e.g., the ones below), and then set this exercise where
the emphasis is on modifications required to get the multi-window case working.
First of all, we will give two simple applications, a calculator and a database,
then we will look at putting them together.
Calculator
In fact, calculator is a bit pretentious, this is really an adder! The user types in
either quit which exits the application, zero which resets the adder to zero or a
number which is added to the running total.
Adder

= add-prompt

!
!

( quit?
skip
[] zero?
show-sum
Adder
show-sum
Adder )
[] number?

The semantic description is given in eventISL.

event: add-prompt =
prompt: true
out: next number
event: quit? =
uses: input
when: input = quit
event: zero? =
uses: input
when: input = zero"
set: sum = 0
out: reset
event: number? =
uses: input, sum
when: is-number(input)
set: sum = sum + input
event: show-sum =
uses: sum
out: sum

June 4, 1993 17:34

exercises

Sheet number 61 Page number 59

Dialogue Notations and Design

Database
Another simple application could be a database, where the user actions are either
quit, set key value, or find key.
Database

Get-key
Get-val

= db-prompt

!
!
!

( quit?
skip
[] set?
Get-key; Get-val
Get-key; print-val )
[] get?
get-key?
= set-key-prompt
= set-val-prompt
get-val?

!
!

We wont bother with the semantic descriptions of the prompt events, or quit
which is the same as for the adder. The descriptions of set? and get? are also
similar to those for quit?, so are omitted.
event: get-key? =
uses: input
set: key = input
event: get-val? =
uses: input, dbase, key
set: dbase = db-update(dbase,key,input)
event: print-val =
uses: dbase, key
out: db-lookup(dbase,key)

The functions db-update and db-lookup are assumed to add a new keyvalue pair and to lookup the value associated with a particular key, respectively.
Multiple windows
Having got these two description, putting them together is quite simple:
Window-system

= Adder || Database

However, it would be considerably more complex if we wanted two adders.

Although this sounds a bit odd, if it were two text editing windows or two spreadsheet windows, this would be quite reasonable. We might attempt to do it simply
as:
Window-system

= Adder || Adder

However, this would not work as the event names and the variables in the
semantic description are global. The windows would both add to the same sum
and it would be random as to which window responded to particular user events.
This is an example of unintentional interference. The only way around this is to
completely copy the description of Adder with new names for most of the events
and variables.

59

June 4, 1993 17:34

60

exercises

Sheet number 62 Page number 60

Dialogue Notations and Design

Students can also be encouraged to look for ways of addressing deliberate
sharing. For example, a common cut-paste buffer can be achieved using shared
variables and explicit dependencies between dialogues can used internal events as
described in section 8.4.3.
The modified dialogue below has an extra command copy in the database,
which is like get except it also puts the value from the database into a shared
buffer and invokes the buffer-ready event, which tells the adder to add the value
in the buffer to its running total.
Database

= db-prompt

!
!
!
!

quit?
skip
Get-key; Get-val
set?
get?
Get-key; print-val
Get-key; Send-val )
copy?
get-key?
= set-key-prompt
= set-val-prompt
get-val?
buffer-ready
= set-buffer

Get-key
Get-val
Send-val

(
[]
[]
[]

!
!

The event copy? is again a simple test of the input and buffer-ready is
described below with the modified adder.
event: set-buffer =
uses: dbase, key
set: val-buffer = db-lookup(dbase,key)
out: db-lookup(dbase,key)

The adder is modified to respond to the event from the database window.
Adder

= add-prompt
( quit?
skip
show-sum
Adder
[] zero?
[] number?
show-sum
Adder )
show-sum
Adder )
[] buffer-ready

!
!

event: buffer-ready =
uses: val-buffer, sum
set: sum = sum + val-buffer

While your students work on this exercise, you can also discuss issues such as
how to represent the changes in keyboard focus. This would probably require an
extra keyboard multiplexer process.

Exercise 8.5
In the example of the digital watch in Section 8.3.8, what would be the
dangerous states? Relate the lexical issues of the buttons for a digital
watch to these dangerous states and provide some design advice. Does
your own digital watch satisfy these criteria?

June 4, 1993 17:34

exercises

Sheet number 63 Page number 61

Dialogue Notations and Design

Answer
The time and alarm setting modes are dangerous states, in that we do not want to
accidentally change either. Design advice would be to put some sort of guard which
makes it difficult to accidentally get into the time setting modes, or alternatively,
once within the modes, to make the buttons which advance the time difficult to
press.
The watch in section 8.3.8 guards these states by requiring button A to be
pressed for two seconds before the mode changes. It might be easy to press a
watch button accidentally, but is quite hard to hold it down. This solution would
not of course work for a keyboard, where it is quite easy to hold a key down
accidentally. Other watches guard these modes by insetting the mode change
button or the buttons required to actually change the time. One has to press these
buttons with a sharp instrument, such as a pencil not an easy slip to make. Some
digital clocks guard the mode by making you hold a button down continuously
while you are changing the time.
Guarding the mode is far preferable to guarding the change buttons, as the
latter makes it very hard to change the time when you want to do it.
In fact changing the time is often a problem for the reversibility of watch and
clock dialogues. Typically there is a button to advance the time by one hour or one
minute, but not one to put the time back. So, if you press the minute advance
button once to often, you have to press it 59 more times to reverse the mistake!

61

June 4, 1993 17:34

exercises

Sheet number 64 Page number 62

June 4, 1993 17:34

exercises

Sheet number 65 Page number 63

Chapter 9

Models of the System

Exercise 9.1
Using the model-oriented approach with the example graphics program
described in Section 9.2.3, specify the move operation as a schema which
acts on the currently selected objects. Is the operation you have defined
cumulative (two successive moves can be done as one move which is the
sum of the other two) or is it forgetful? Discuss the implications of the
framing problem in your definition.

Answer
Students may not have a great familiarity with the Z notation used in this textbook,
especially if they do not have a computer science background which emphasizes
formal specification. Nevertheless, we will provide a model answer to this exercise
in the Z notation and leave it to the instructor to decide whether an alternative
language can be used to express the same ideas.
To accommodate the specification of the move operation, we will first need a
function which allows us to add two coordinate points together. We just provide
the declaration of this function here.
+ : Point  Point ! Point
The motivation for this function is so that we can define a move operation with a
single argument, called delta? below, which will specify how far to move the centres
of every selected shape. This approach, using relative movements, will allow for
two consecutive moves to be defined with the overall movement being the sum
of the individual moves. Another approach (which would be quite cumbersome
to define for a set of selected objects) would be to specify the move operation as
an absolute operation, the argument(s) to the move explicitly indicating the new
value for the centre(s) of each selected shape. This operation definition would not
be cumulative.
63

June 4, 1993 17:34

64

exercises

Sheet number 66 Page number 64

Models of the System

Initially, we might want to define the move operation simply to update the
centre position of all selected shapes.
Move
State
State0
delta? : Point

8 id : selected  shapes (id) centre = shapes(id) centre + delta?

0

There are a number of details in this defined operation which are left (dangerously) open to interpretation. Many of these are framing conditions on the
Move operation. For example, what happens to shapes which are not selected?
The normal intuition on this operation would be that they stay put, but the above
definition actually allows them to change in any way including the possibility
of disappearing altogether. To ensure that unselected objects stay the same as they
were before the operation, we would have to add a predicate to the definition.
Move
State
State0
delta? : Point

8 id : selected  shapes (id) centre = shapes(id) centre + delta?

8 id : (dom shapes , selected)  shapes (id) = shapes(id)
0

We also need to explicitly state that no new shapes are introduced into the
system, i.e., that the domain of shapes remains the same.
Move
State
State0
delta? : Point

8 id : selected  shapes (id) centre = shapes(id) centre + delta?

8 id : (dom shapes , selected)  shapes (id) = shapes(id)
0

dom shapes0 = dom shapes

For those familiar with Z, it is possible to collapse the last three predicates in
the above schema to one predicate using functional overriding from the basic Z
mathematical toolkit (see Spivey, J.M., The Z Notation: A Reference Manual, Prentice
Hall).
Another framing issue which has impact on the interactive style of this graphics
system is specifying what happens to the selected set of shapes after the move
operation. Our current definition says nothing about this. Two clear options are

June 4, 1993 17:34

exercises

Sheet number 67 Page number 65

Models of the System

whether the set of selected shapes remains the same or becomes the empty set (no
selected objects). For reasons of compositionality (for defining consecutive move
operations or consecutive operations targeted at the same set of shapes) we choose
the former option, and specify this below.
Move
State
State0
delta? : Point

8 id : selected  shapes (id) centre = shapes(id) centre + delta?

8 id : (dom shapes , selected)  shapes (id) = shapes(id)
0

dom shapes0 = dom shapes

selected0 = selected
[Note: Experienced Z users would probably choose the functional overriding
operator () to more concisely express the first three predicates above.]

Exercise 9.2
Write a similar schema for RESIZE. It should have the following informal
semantics. The width and height attributes are of the shapes bounding
box. The resize operation should make one corner of the bounding box be
at the current mouse position (supplied as the argument current pos?).
Hint: the width of a box is twice the difference between the x coordinate
of the centre and the x coordinate of any corner:
wid = 2 j centre:x , corner:x j

Answer
Strictly speaking, there are some errors in the Z usage for the hint in this exercise.
The type Point is not a schema type, and so it is not meaningful to use schema
selection, as in expressions like centre:x and corner:x. The alert Z student should
pick up on this error. However, spirit of the hint is still valid. To make things legal
in Z, we can change the declaration of Point as follows.
Point
x; y : R
To best specify this resizing operation, we will require to explicitly model
the bounding box of a shape. We introduce a type, called BoundingBox which

65

June 4, 1993 17:34

66

exercises

Sheet number 68 Page number 66

Models of the System

will represent the corners of the surrounding box which defines the extent of a
shape. In fact, in Figure 9.1,we have drawn the bounding box for the circle but
we have not made that information explicit in the specification in the text of this
chapter. We omit details in this sample answer on rectangular constraints for
the corners of the bounding box (e.g., that upper left and right corners have the
same y-coordinate values but the x-coordinate of the upper left corner is strictly
less than the x-coordinate of the upper right corner, etc.
BoundingBox
upperleft; upperright;
lowerleft; lowerright : Point
upperleft:x  upperright:x
upperleft:y = upperleft:y
:::

We can add the bounding box as information known about shapes, by extending the definition of Shape. A constraint on the bounding box, motivated by the
hint to this exercise, is that its width and height are the same as the width and
height of the shape. In addition, the centre coordinates are related to the width
and height. Note below that we use the names height and width in the definition of
Shape to increase its readability.
Shape
type : Shape type
width : R
height : R
centre : Point
bb : BoundingBox
width = 2  (centre:x , bb:upperleft:x)
height = 2  (centre:y , bb:upperleft:y)
width = (bb:upperleft:x , bb:upperright:x)
height = (bb:upperleft:y , bb:lowerleft:y)

Resizing can be viewed as changing the size of the bounding box. At a relatively
abstract level, we can simply specify the resizing operation with a single argument
indicating the new bounding box for the selected shape. A precondition for the
operation is that there is only one selected shape.

June 4, 1993 17:34

exercises

Sheet number 69 Page number 67

Models of the System

Resize
State
State0
newbb? : BoundingBox
#selected = 1
8 id : selected  shapes0 (id):bb = newbb?
forallid : (dom shapes , selected)  shapes0 (id) = shapes(id)
dom shapes0 = dom shapes
selected0 = selected

Exercise 9.3
In Section 9.2.5, we said that the specification Algebraic , Zdraw could be
extended to say that a move, resize or unselect after a delete has no effect.
The axiom for unselect looks like this:
(ix) unselect(delete(g)) = delete(g)
Write two more axioms (x) and (xi) which say the same about move and
resize. Now show use axioms (iv) and (v to show that (ix) implies both
your new axioms.

Answer
The three additional axioms are:
(9) unselect(delete(st)) = delete(st)
(10) move(p; delete(st)) = delete(st)
(11) resize(p; delete(st)) = delete(st)
Axioms 4 and 5 are:
(4) move(p; unselect(st)) = unselect(g)
(5) resize(p; unselect(st)) = unselect(g)
To show that Axiom 10 is derivable from Axioms 4 and 9 we can argue as
follows:
move(p; delete(st))
= move(p; unselect(delete(st))
= unselect(delete(st))
= delete(st)

by Axiom 9
by Axiom 4
by Axiom 9

A similar argument shows that Axiom 11 is derivable from Axioms 5 and 9.

67

June 4, 1993 17:34

68

exercises

Sheet number 70 Page number 68

Models of the System

Exercise 9.4

A two-function calculator has the following buttons 09, +, - and =. These

would comprise the command set (C) of a PIE model of the calculator. The
state (E) would consist of three components: the two current number being
entered, the last number (to be operated on) and the pending operation (+
or -). The display (D) is simply a signed number. For this example, ignore
the result.
Write down (semi-formally) the doit function updating the state for each
user command and the display function relating the state (E) to the current
display (D). To check your definitions: what does the display have on it
after the user has entered 2+3+? Most calculators would show 5, does
yours?
Consider the displays after the sequences 2+2 2+ and the effect on
each of the additional user input of 3. Does the calculator satisfy the
transparency property?

Answer
The definition of the doit function could proceed as follows. We will use the Z
notation in this sample answer. The character set is defined as:
C ::= 0 j 1 j 2 j 3 j 4 j 5 j 6 j 7 j 8 j 9 j 0 j + j , j=
It will be useful to distinguish the digits and operations (+ and ,).
Digit : P C
Op : P C

Digit = f 0; 1; 2; 3; 4; 5; 6; 7; 8; 9; 0 g
Op = f +; , g
The state is defined as an ordered four-tuple, consisting of the current number
being entered,the last number (to be operated on) and the pending operation.
E == seq Digit  seq Digit  Op
The display is simply a projection of the current number being entered.
D == seq Digit

June 4, 1993 17:34

exercises

Sheet number 71 Page number 69

Models of the System

display : E ! D

8(current last op) : E 

;

display(current; last; op) = current

The doit function is defined below by case exhaustion. The notation o(c; l)
means to apply operation o to the arguments c and l, so that 2 + 3 would be written
+(2; 3), with o = +, c = 2 and l = 3.
doit : E  C ! E

8 c l : seq Digit; d d
;

: Digit; o; o0 : Op 
doit((c; l; o); =) = (o(c; l); 0; o)
if lastkey 2 f+; ,; =g
doit((c; l; o); d) = (d; c; o)
(10c + d; l; o) otherwise
doit((c; l; o); o0) = (o(c; l); c; o0)
;

The following table traces the execution of the calculator.

C
2
+
3
+

E
(0,0,+)
(2,0,+)
(2,2,+)
(3,2,+)
(5,3,+)

D
0
2
2
3
5

Compare the trace tables for the programs 2 + 3 and 2 + 23.

C
2
+
3

E
(0,0,+)
(2,0,+)
(2,2,+)
(3,2,+)

D
0
2
2
3

C
2
+
2
3

E
(0,0,+)
(2,0,+)
(2,2,+)
(2,2,+)
(23,2,+)

D
0
2
2
2
23

Notice that the display after the program 2+ is the same as it is after the program
2 + 2, yet the behaviour of the subsequent 3 command is different. Therefore, the
calculator as defined does not satisfy the transparency property.

Exercise 9.5
Imagine a calculator, normal except it displays A for 0, B for 1, up to J for
9. So the number 372 would appear as DHC. Does this affect the formal
transparency of the calculator? Should it?

69

June 4, 1993 17:34

70

exercises

Sheet number 72 Page number 70

Models of the System

Answer
Formally, this would not affect the transparency property of a calculator. Any
function from display to effect which held for the numbers would also hold for
their alphabetic counterparts by just composing it with the code A for 0, B for
1, up to J for 9. Of course, most people would probably find using the encoded
calculator quite a bit more difficult than one which used the normal ten digits. The
important lesson here is that the formal expression of the transparency property
does not indicate how difficult it is for a human to decide determine the function
between display and effect. The formal expression just requires that there be some
function. In practice, human users would probably judge the encoded calculator
as unpredictable (that is, not transparent). The moral is that even though we
can capture a usability property formally does not mean that we have captured
the entire intent of the property. The transparency property would have to be
augmented in order to measure the complexity from the users perspective of the
function between display and effect.

Exercise 9.6
Can you suggest any improvements to the screen button feedback problem
discussed in Section 9.4 that would distinguish at the interface between
the two cases of hitting or missing the button? Is there any guarantee with
your solution that the user will notice the distinction?

Answer
Some suggested fixes to the button feedback problem would be to have the function
attached to the button be invoked when the mousekey is pressed down, rather than
up. But this solution would not work for functions invoked on pop-up menus.
Another possibility is to have some visible or audible feedback from the button
associated with the invocation of the function on the mousekey release. This
solution does not guarantee that the user notices the added effect unless their
attention is focused on the added visual or aural effect. Another possibility, subtly
different from the last suggestion, would be to associate the addition verbal or aural
cue to the error case, when the mouse accidentally slips off the button between the
press and release of the mousekey.

June 4, 1993 17:34

exercises

Sheet number 73 Page number 71

Chapter 10

Implementation Support

Exercise 10.1
In contrasting the readevaluation loop and the notification based paradigm for interactive programs, construction of a preemptive dialogue was
discussed. How would a programmer describe a preemptive dialogue by
purely graphical means?

Answer
Creating a preemptive dialogue by graphical means is quite simple. In a graphicbased interaction, it is assumed that the user can only interact with parts of the
system that are visible. In a windowing system, for example, the user can only
direct input to a single, currently selected window and the only means of changing
the selected window would be by indicating with some gesture within that window. To create a preemptive dialogue, the system can produce a window which
covers the entire screen and make it the currently selected window. All user input
would then be directed to that window and the user would have no means of
selecting another window. The covering window thus preempts any other user
action except that which it is defined to support.

Exercise 10.2
Recall the state transition diagram for font characteristics presented in
Chapter 8 (Section 8.3.3). Compare different interaction objects which
could implement this kind of dialogue. Use examples from existing
toolkits (pull-down menus or dialogue boxes) or create a novel interaction object.
71

June 4, 1993 17:34

72

exercises

Sheet number 74 Page number 72

Implementation Support

Answer
To recall the example, we want to allow the user to toggle between three font
characteristicsbold, italics and underline. Pull-down menus and dialogue boxes
are the most frequently used methods for implementing such a dialogue. A pulldown menu would have the three choices listed with some way of indicating
whether the characteristic was active or not. Upon each invocation of the menu,
the user could change the active status of one font characteristic. For example, they
could turn on the bold feature or turn off the underline feature. The important
feature of this dialogue is that more complex tasks (changing the active status of
two or more font characteristics) can only be achieved by separate invocations of
the menu. And often the only way the user can check the status would be by
popping up the menu again (without invoking any action). This last point shows
that the menu design alone does not support immediate display of relevant state
changes as discussed in Chapter 4 under the synthesizability principle.
Within a dialogue box, the user can affect several changes at one invocation.
Technically, it is not the dialogue box interaction device which allows for this, but
the use of toggle buttons. Each font characteristic is represented by a box which
acts as a simple on/off switch for that characteristic. As an aid to the user, the
toggle button can change its appearance so that the user can know its status (on
or off) by looking at it. A similar interaction widget is the checkbox, which would
be labelled with the particular font characteristic. Clicking on the checkbox would
put a mark in the box to indicate that the characteristic is on. Another click would
remove the mark. In fact, a dialogue box need not be used to portray the set of
toggle buttons or checkboxes, as they could be presented continually so that the
user can always view the current font characteristic settings. The reason a dialogue
box might be used is because there may be many font characteristics to choose from
and the dialogue box will keep them grouped together and preserve screen space
by allowing them to be hidden and used on an as-needed basis.

Exercise 10.3
Scrolling is an effective means of browsing through a document in a window that is too small to show the whole document. Compare the different
interactive behaviour of the following two interaction objects to implement
scrolling:

A scroll bar is attached to the side of the window with arrows at

the top and bottom. When the mouse is positioned over the arrow
at the top of the screen (which points up), the window frame is
moved upwards to reveal a part of the document above/before what
is currently viewed. When the bottom arrow is selected, the frame
moves down to reveal the document below/after the current view.

June 4, 1993 17:34

exercises

Sheet number 75 Page number 73

Implementation Support

The document is contained in a textual interaction object. Pressing

the mouse button in the text object allows you to drag the document
within the window boundaries. You drag up to browse down in the
document and you drag down to browse up.

The difference between the two situations can be characterized by noticing

that in the first case, the user is actually manipulating the window (moving
it up or down to reveal the contents of the document), whereas in the
second case, the user is manipulating the document (pushing it up or down
to reveal its contents through the windows. What usability principles
would you use to justify one method over the other (also consider the case
when you want to scroll from side to side as well as up and down)? What
implementation considerations are important?

Answer
There are many usability principles which can be brought to bear on an examination
of scrolling principles. For example:
Observability The whole reason why scrolling is used is because there is too
much information to present all at once. Providing a means of viewing document contents without changing the contents increases the observability of
the system. Scrollbars also increase observability because they help to indicate
the wider context of the information which is currently visible, typically by
showing where the window of information fits within the whole document.
However, observability does not address the particular design options put
forth here.
Predictability The value of a scrolling mechanism lies in the user being able to
know where a particular scrolling action will lead in the document. The
use of arrows on the scrollbar are used to help the user predict the effect
of the scrolling operation. If an arrow points up, the question is whether
that indicates the direction the window is being moved (the first case) or the
direction the actual text would have to move (the second case). The empirical
question here is to what object do users associate the arrow, the text or the
text window? The arrow of the scroll bar is more closely connected to the
boundary of a text window, so the more usual interpretation would be to have
it indicate the direction of the window movement.
Synthesizabilty You might think that it does not matter which object the user
associates to the arrow. They will just have to learn the mapping and live with
it. In this case, how easy is it to learn the mapping, i.e., can the user synthesize
the meaning of the scrolling actions from changes made at the display? Usually,
the movement of a box within the scrollbar itself will indicate the result of a
scrolling operation.

73

June 4, 1993 17:34

74

exercises

Sheet number 76 Page number 74

Implementation Support
Familiarity/Guessability It would be an interesting experiment to see whether
there was a difference in the performance of new users for the difference
scrolling mechanisms. This might be the subject of a more extended exercise.
Task conformance There are some implementation limitations for these scrolling
mechanisms (see below). In light of those limitations, does the particular
scrolling task prefer one over the other. In considering this principle, we need
to know what kinds of scrolling activity will be necessary. Is the document
a long text that will be browsed from end to end, or is it possibly a map or
a picture which is only slightly larger than the actual screen so scrolling will
only be done in small increments?
Some implementation considerations:





What scroll mechanisms does a toolkit provide? Is it easy to access the two
options discussed above within the same toolkit?
In the case of the second scrolling option, are there enough keys on the mouse
to allow this operation without interfering with other important mouse operations, such as arbitrarily moving the insertion point or selecting a portion of
text or selecting a graphical item?
In the second option, the user places the mouse on a specific location within the
window, and gestures to dictate the movement of the underlying document.
What kind of behaviour is expected when the mouse hits the boundary of the
window? Is the scrolling limited in this case to steps bounded in size by the
size of the window, so that scrolling between two distant points requires many
separate smaller scrolling actions?

June 4, 1993 17:34

exercises

Sheet number 77 Page number 75

Chapter 11

Evaluation Techniques

Exercise 11.1
In groups or pairs, use the cognitive walkthrough example, and what you
know about user psychology (see Chapter 1), to discuss the design of a
computer application of your choice (for example, a word-processor or
a drawing package). Hint: focus your discussion on one or two specific
tasks within the application.

Answer
This exercise is intended to give the students a feel for using the technique of
cognitive walkthrough (CW). CW is described in detail in the chapter and the
same format can be used here. It is important to focus on a task that is not too
trivial, for example, creating a style in a WP package. Also assume a user who
is familiar with the notion of styles (and with applications on the same platform
(e.g. Macs, PCs, UNIX etc.) but not with the particular word processing package.
Attention should be given to instances where the interface fails to support the user
in resolving the goal and where it presents false avenues.

Exercise 11.2
Design an experiment to test whether adding colour coding to an interface
will improve accuracy.

Answer
The following is only an example of the type of experiment that might be devised.
Subjects taken from user population
75

June 4, 1993 17:34

76

exercises

Sheet number 78 Page number 76

Evaluation Techniques
Hypothesis colour coding will make selection more accurate
IV (independent variable) colour coding
DV (dependent variable) accuracy measured as number of errors
Design between groups to ensure no transfer of learning (or within groups with
appropriate safeguards if subjects are scarce).
Task The interfaces are identical in each of the conditions, except that, in the
second, colour is added to indicate related menu items. Subject is presented
with a screen of menu choices (ordered randomly) and verbally told what they
have to select. Selection must be done within a strict time limit when the screen
clears. Failure to select the correct item is deemed an error. Each presentation
places items in new positions. Subjects perform in one of the two conditions.
Analysis t test

Exercise 11.3
You have been asked to compare user performance and preferences with
two different learning systems, one using hypermedia (see Chapter 15),
the other sequential lessons. Design a questionnaire to find out what the
users think of the system. How would you go about comparing user
performance with these two systems?

Answer
Assume that all users have used both systems.

Questionnaire
Consider the following questions in designing the questionnaire:




what information is required?

how is the questionnaire to be analyzed?

You are particularly interested in user preferences so questions should focus on

different aspects of the systems and try to measure levels of satisfaction. The use
of scales will make responses for each system easier to compare.
Table 11.x1 shows an example questionnaire.
To test performance you would design an experiment where two groups of
subjects learn the same material using the two systems, and test how well they
have learned (using a standard measurable test).

June 4, 1993 17:34

exercises

Sheet number 79 Page number 77

Evaluation Techniques

PART I: Repeat for each system

Indicate your agreement or disagreement with the following statements.
(1 indicates complete disagreement and 5 complete agreement).
The system tells me what to do at every point.
Disagree 1 2 3 4 5 Agree
It is easy to recover from mistakes.
Disagree 1 2 3 4 5 Agree
It is easy to get help when needed.
Disagree 1 2 3 4 5 Agree
I always know what the system is doing.
Disagree 1 2 3 4 5 Agree
I always know where I am in the training material.
Disagree 1 2 3 4 5 Agree
I have learned the material well using the system.
Disagree 1 2 3 4 5 Agree
I could have learned the material more effectively using a book.
Disagree 1 2 3 4 5 Agree
I always know how well I am doing.
Disagree 1 2 3 4 5 Agree
PART II: Comparing both systems:
Which system (choose 1) was most
helpful
A B
effective A B
usable
A B
Please add any comments you have about either system:

Table 11.x1 Questionnaire to compare two systems

77

June 4, 1993 17:34

78

exercises

Sheet number 80 Page number 78

Evaluation Techniques
Subjects user group
IV style of learning system
DV performance (measured as test score).
Design Between groups design.

Exercise 11.4
What are the benefits and problems of using video in experimentation? If
you have access to a video recorder, attempt to transcribe a piece of action
and conversation (it does not have to be an experiment a soap opera
will do! ). What problems did you encounter?

Answer
The benefits of video include: accurate realistic representation of task performance
especially where more than one video is used; a permanent record of the observed
behaviour.
The disadvantages include: vast amounts of data which are difficult to analyze
effectively; transcription; obtrusiveness; special equipment required.
The student can try transcribing a video record conversation from a TV
soap will do. The aim is to get the student to experience some of the difficulties of
representing a visual record in a semi-formal written format. The students should
discuss which parts of the video are most difficult to represent, and how important
these parts are to understanding the clip.

Exercise 11.5
In Section 11.5.1, we saw that the observed results could be the result of
interference. Can you think of alternative designs that may make this less
likely. Remember that individual variation was very high, so you must
retain a within-groups design, but you may perform more tests on each
subject.

Answer
The students should read carefully the description of the experiment in Section 11.5.1 to look for ways of reducing the possible interference. Three possibilities
are:

June 4, 1993 17:34

exercises

Sheet number 81 Page number 79

Evaluation Techniques

During the initial training period swop back and forth between learning the
two sets of icons, with the aim of getting the subjects used to swopping between
the two sets of remembered icons. However, this design could be argued to
suffer the same flaws as the original. If the abstract icons had been taught in
isolation perhaps they might have fared far better.
One could invent a third set of random icons (call them R). We could then
interpose them in the experiment, that is present the icons in the orders RARN
and RNRA. The intention is to swamp any transfer effect in the noise of
the random icons. It could be argued that our experiment then measures the
robustness of the icon sets to such noise!
We could give the subjects multiple presentations, for example, ANAN and
NANA presentation orders. This would not remove transfer effects, but does
give us some way to quantify them. Imagine that in the ANAN group that
the second presentation of the abstract icons was significantly worse than the
first, but there was not a similar effect for natural icons in the NANA group.
This would both give us positive evidence of a transfer effect, and perhaps
give us some quantitative measure. However, even going from this additional
evidence to a strong conclusion will be difficult.

Notice that all the above measures require additional subject time and one has to
constantly weigh up the advantages of richer experiments against those of larger
subject groups.

79

June 4, 1993 17:34

exercises

Sheet number 82 Page number 80

June 4, 1993 17:34

exercises

Sheet number 83 Page number 81

Chapter 12

Help and Documentation

Exercise 12.1

Write a manual page for making a cup of coffee. Assume your user has
no experience but will recognize a cup, a kettle, a spoon, etc. Swap your
manual with a partner. Does your partners manual give you sufficient
instruction to make the cup of coffee? Discuss improvements with your
partner and agree on a final version of the manual.

Answer
First you have to decide upon the level of granularity which you are going to work
at. The aim of the exercise is to demonstrate that it is not as straightforward as it
might seem to provide instructions even for a very familiar and well understood
task. Given this, it is most helpful to assume that the user knows very little.
The example solution (Table 12.x1) assumes that the user will recognize objects
(perhaps they are labelled) and understands common actions and directions, but
not the specific actions required here. Alternatively you could choose to assume
that the user does know how to turn on a tap and open a jar.
N.B. The two options could be expanded further if required. Other alternatives
could be included such as getting water from another source.
The discussion should focus on the assumptions that are made. The student
should make a conscious decision as to what assumptions to make, rather than
making them by default. This should ensure that the assumptions that are made
are appropriate to the particular user.
The students could also contrast their answers here, based on the material in
this chapter, with the making tea manual in Chapter 7. How useful do they feel
task analysis is in designing a manual?
81

June 4, 1993 17:34

82

exercises

Sheet number 84 Page number 82

Help and Documentation

Manual for making a cup of coffee

Required:

an automatic electric kettle, a jar of instant coffee powder or

granules, a mug, a teaspoon, a mains water tap (or an alternative
source of water), a mains electricity supply, milk (optional), sugar
(optional).

CAUTION:

Electricity can be dangerous avoid any contact between electric connections and water.
Boiling water can scald take care.

To boil water:
Ensure plug on kettle is not connected to mains electricity supply.
Remove lid from kettle.
Place kettle directly beneath spout of tap.
Turn tap handle anticlockwise to release water.
When water reaches mark labelled full on kettle turn tap handle
clockwise to close off water supply.
Replace kettle lid.
Move kettle to proximity of mains electricity supply.
Place kettles plug into electricity socket.
Press button marked on on top of kettle to switch kettle on.
When the water has boiled, the kettle will switch off automatically and the on button will return to its original position with a
click.
Remove kettles plug from electricity socket.

To make coffee:
Take jar of coffee and remove lid by turning anticlockwise.
Fill teaspoon with coffee.
Place contents of teaspoon in cup.
Replace lid on coffee jar and turn clockwise to tighten.
Pour boiling water from the kettle into the cup up to approximately 1/2 an inch from the top.
Add milk if required (to almost fill cup).
Add sugar if required.
Stir coffee mixture with spoon.
The coffee is now ready to drink.

Table 12.x1 Coffee making manual

June 4, 1993 17:34

exercises

Sheet number 85 Page number 83

Help and Documentation

Exercise 12.2
Find a computer application that you have never used before. Attempt to
learn to use it using only the on-line support. Is there enough information
to allow you to use the application effectively? Is the information easy to
find? What improvements (if any) would you suggest?

Answer
This is an investigative exercise for which there is no example solution. Possible
systems to consider would be a word processor (for example, Word, WordPerfect,
Word for Windows), a graphics package (MacDraw, Superpaint, CorelDraw) or
a spreadsheet (Excel, Lotus). It may be helpful to provide the student with a
list of tasks to learn to perform using the application. Include in this list basic
tasks (creating a document, spreadsheet etc.) and more complex ones (creating
templates, etc.). The aim is to get the student to think about the provision of help
from the point of view of the solitary user trying to figure things out alone. The
student should therefore make explicit any prior knowledge used to help interpret
the system (for example, use of similar systems).

Exercise 12.3
What knowledge is needed to build an intelligent help system? Which do
you think is most difficult to provide and why?

Answer
It requires knowledge of the domain and knowledge of the user at least. It may
also require knowledge of teaching strategies and tasks. Knowledge of the user
is usually most difficult to provide along with knowledge of the task. Even if the
user can be monitored, interpreting the users behaviour in anything other than
coarse terms is difficult without access to his thought processes. However this level
of interpretation, and generalization, can be helpful nonetheless (for example, a
record of what the user has already done successfully or how often he has used
the application). Context can be deduced from the users current activity. This too
can be difficult to determine however, unless it is a simple case where the tool in
use indicates the activity being performed.
Domain knowledge in the general sense is probably the easiest to provide
(although it is time consuming to do so). It is available within the system itself or
from the designer.

83

June 4, 1993 17:34

84

exercises

Sheet number 86 Page number 84

Help and Documentation

Exercise 12.4
Look at as many on-line support systems as you can. Which do you find
most useful and why? Try to assess them using the requirements discussed
in Section 12.2.

Answer
This is another open investigative assignment. The set of requirements in Section
12.2. should provide the student with a structure for evaluating the systems. A
scoring system can perhaps be worked out to aid comparison. Try to compare a
general context sensitive help system (for example, Balloons help) with an application specific one. Also a hypertext based help system would provide an interesting
contrast as would a manual based help system (such as UNIX man) and a prompt
based one.

June 4, 1993 17:34

exercises

Sheet number 87 Page number 85

Chapter 13

Groupware

Exercise 13.1
We discussed the use of a group pointer in a shared editor with a shared
view. Consider the advantages and problems of using a group pointer
when participants have different views. How do you show the pointer if
it is outside part of the document you are working on? Think also about
the issues when the system is a hypertext based co-authoring system. Is
there any use for a group pointer in this case?

Answer
The purpose of this exercise is to encourage the students to think of the various
scenarios which can arise during group editing and to encourage their creative
interface thinking. Where suitable tools are available, the students can be encouraged to create mock-ups of their solution using a drawing package, or even
envisionments of specific scenarios using HyperCard or a presentation tool. Below
are some ideas.
We can start off with the shared editor shown in Figure 13.7. Imagine if your
colleague wanted to point to the line starting Thinking about. She could select
the group pointer tool at which point she can drag around an icon shaped like
a hand. She moves this over the words Thinking about and starts to talk about
the text (using some other means of direct communication such as the telephone).
As the phrase in question is on both displays there is no problem about where to
display the hand, it is placed over the words Thinking about on your display.
(Note that the group pointer is at a different physical location on each display).
Arguably the group pointer is even more important when there are separate
views then when the view is shared. With a shared view the participants can be
assured that they are seeing the same thing and so can assume a shared context.
With separate views the participants are likely to have various problems when
they forget that the others views are different. For example, they will have trouble
85

June 4, 1993 17:34

86

exercises

Sheet number 88 Page number 86

Groupware
with indexical expressions or with relative statements the thing at the top. The
presence of a group pointer can help the participants to gain a shared focus.
On the other hand, the very presence of the group pointer can encourage the
participants in a false sense of shared context. One could therefore argue that when
participants want to enter a period of shared focus and synchronous discussion
that they also move to a shared view. However, for the rest of this answer well
assume that the views stay different.
What do we do when the group pointer is outside of the view? For example,
in Figure 13.7, your colleague might move the group pointer to the line starting
More adaptable. In this case, your screen cannot display the group pointer in its
correct position. Various options suggest themselves.
One could have the hand icon appear at the very top of your display with
the finger pointing straight upward, meaning see above. Similarly if the group
pointer were below the screen the icon could appear at the bottom pointing down.
If such icons were displayed then the user can scroll the screen upwards until the
group focus was found. Alternatively, one could arrange that clicking on the icon
scrolled one to the correct position.
Instead or in addition to these out of view icons, one could use the scroll bar
and display the group pointers position in it, thus allowing the user to easily see
where it is and scroll there if appropriate.
In the hypertext case, the solutions depend very much on the particular style
of the system. If the individual nodes are large, then even if two participants
are acting on the same node they may have different views, in which case all the
above measures become relevant. If the participants are looking at different nodes,
then the first action of a group pointer should be to direct attention to the node of
interest. In a hypertext where the node graph is shown, this can be achieved by
displaying the group pointer on the node of interest. Alternatively, if the graph is
not displayed, the presence of a group pointer could be shown by adding a go to
the focus node button, rather like the home button in many HyperCard stacks.
Although it is not mentioned in the question, it is worth also considering the
display of other peoples insertion points. In Figure 13.7, although your insertion
point is within your colleagues view, it is not displayed in any way. There is
no reason why other peoples insertion points should not be displayed (suitably
identified) when they fall within your view and visa versa. In addition, one might
want some indicator on your scroll bar to show where your colleagues current
view is. Such indicators can enhance group awareness and be a focus for initiating
direct communication.

Exercise 13.2
Find out how many different forms of direct computer-mediated communication are available on your system (start with email). Are they heavily

June 4, 1993 17:34

exercises

Sheet number 89 Page number 87

Groupware
used, and if so, where do they fit in the time/space matrix (Figure 13.8)
and its refinement (Figure 13.9).

Answer
Obviously the answers to this and Exercise 13.3 depend on your own particular
facilities, but here are some suggestions based on the authors sites.
The question says starting with email. In fact, on many systems several email
interfaces are available. On one of the authors system there are two command line
interfaces as well as various GUI interfaces for different window managers (under
X, on the Macintosh etc.). Although email is just one communication mechanism,
it is used in a wide variety of ways. Its normal use is essentially asynchronous/remote, or in terms of the refined matrix unsynchronised. However, delivery
over a LAN is often virtually instantaneous and you can observe users engaging in
effectively synchronous communication. The possibility of these modes of use is often determined by the style of the interface as is shown in section 9.4.4. Clearly, the
way in which new mail is notified and the speed with which mail can be read both
determine the pace at which the conversation can proceed. A further complication
in some sites is that internal and external email have entirely different interfaces:
sending mail off site may involve remotely logging into a gateway machine and
then using some obscure mailer there, rather than the familiar workstation based
tools.
In addition to email, you may have one or more text based synchronous communication tools. For example, many UNIX machines have two tools write
and talk. The command write Alison followed by some text prints that text
on Alisons screen. She can respond by doing a write back to the sender. The
talk command establishes a more continuous conversation, as described in section 13.3.1, where both participants screens are split in two, each half displaying
one of the participants contributions. Although, both can be seen as synchronous/remote, they operate at different levels of sharing (section 13.6.2). Whereas
write only sends the text after the contribution is completed, talk sends each
users contributions character by character as they are typed. That is talk sits
further down towards the bottom left of the diagram in Figure 13.10. Note also
that write can be used in a semi-asynchronous mode. If the other user is logged
in, but not at their terminal, the message waits there until their return.
Another asynchronous communication tool commonly available is bulletin
boards, such as Usenet news. As discussed in Section 13.3.1, these have different
properties to email, being rather less personal.
In addition, to these standard tools, some of the authors colleagues have
permanent video connections between their offices. You may have similar facilities,
or other computer augmented communications, perhaps video in a window on
your workstation. Dont forget apparently low-tech things such as computers
which automatically dial telephone calls.

87

June 4, 1993 17:34

88

exercises

Sheet number 90 Page number 88

Groupware

Exercise 13.3
Repeat the previous exercise, but this time look for shared data on your
system. Is the data updated by one person and viewed by many, or have
you files or databases which are updated by several people? If the latter,
find out what methods are used to prevent two users changing the same
data at the same time. There may be no mechanism at all, a computerised
one (e.g., locking) or a social protocol (e.g., a floppy disk is passed around).

Answer
There is an argument that bulletin boards are a form of shared data, so should
certainly be acceptable as an answer to this question, depending on their nature at
your site and also on how well argued the answer is.
You are likely to come across a lot of loosely shared items, where only a small
group access them. For example, the files which comprise the book were shared
between the authors. Whether any sort of locking is used will depend on the
facilities available in your installation, but typically such small group sharing is
handled by social protocols (sometimes unsuccessfully!).
Here are some more specific examples from one of the authors sites:
There are centralised databases containing phone numbers and car numbers.
The former is used in place of the internal telephone directory and the latter when
you notice some lights have been left on in the car park! Both of these are updated
centrally by one of the administrative staff. Changes are notified (by email) to the
administrator. However, access to the information is available to all.
Each user has a plan file in their local directory into which she can put generally informative remarks about her whereabouts. Typically, this might include
a contact telephone number, regular appointments a simple diary, or indeed anything of interest. When another user enters the command finger Alison, Alisons
plan file is displayed. As the file is owned by the individual user, normal access
permissions apply and so typically a user can only update their own plan file.
However, through the finger command the information is publically available.
There are also examples of publically updatable shared files. For example,
some email groups are handled by having a file containing the email address of
each recipient. Anyone can edit the file to add or remove their name from this list.
(In fact, they can just as easily add or remove someone elses name!) On our system
this file has no form of protection from simultaneous update, the assumption being
that such updates are so rare that they will never happen (really).
The lack of shared data with locking is a reflection of the poor level of system
support in the UNIX world. Users of PC based software should be able to find
examples of locking of shared files or databases.
Finally, it is worth discussing the use of implicitly shared information. For
example, the finger command in UNIX also tells you whether the user is logged
in and, if so, where. This is information known to the system, but made available
to the users in such a way as to facilitate cooperation.

June 4, 1993 17:34

exercises

Sheet number 91 Page number 89

Chapter 14

CSCW Issues and Theory

These are both mini-project exercises, so do not have definitive answer. here are
some pointers to look out for.

Exercise 14.1
In Section 14.3.2, we discussed the highly contextual nature of the spoken
word, including the use of deictic reference and indexicals, and the (officially) ungrammatical and fragmentary use of sentences. Try listening
to social chat over cups of tea collect examples of different forms of
contextual utterance.

Answer
There are often objects of shared attention during social chat, perhaps a notice
board, or magazines on a coffee table. These will be one obvious source of context.
Make sure that when the students record statements such as I like that, they
also record whether the indexical refers to a previous utterance or to something
external. Encourage them to observe exactly how external references are made: for
example, whether people point to a notice, refer implicitly to the article someone
else is reading or use the direction of their own gaze.
In a small group, the students should listen particularly carefully for the use
of the pronoun you. Does it refer to the whole group (excluding the speaker), or
to a particular person. If the latter, then how is it made explicit eye gaze, body
position? Other pronouns like we or she can be be equally contextual.
The students should also look out for implicit context, for example, a longer
tail would improve the aerodynamics, spoken when the listener is looking at a
picture of a kite. Although the sentence has no pronoun there is an implicit subject
of the kite youre looking at. This form of contextual statement can become
arbitrarily obscure, for example, as one person meets another: its in your pigeon
hole referring to an overdue report.
89

June 4, 1993 17:34

90

exercises

Sheet number 92 Page number 90

CSCW Issues and Theory

As the students become more adept they can listen more carefully looking
for the shift in conversational focus, and breakdowns where the context is not
successfully negotiated.

Exercise 14.2
Go into an office or other place where several people are working together.
Try to note down in as much detail as possible what they are doing and
when. Do this with different foci: focus on the direct interpersonal communications, focus on the shared objects such as a calendar or document,
or focus on one worker at a time. Whilst collecting data and when ordering
your notes, look for breakdowns and misunderstandings, and for implicit communication through objects. Look also at a particular task over a
period of time, and note the number of interruptions as a worker performs
the task, or the way a single task is contributed to by several workers.

Answer
This exercise is similar to Exercise 7.5. However, whereas the task analysis in
7.5 is quite structured this exercise is not intended to produce the same form of
precise task sequences etc. The intention is to expose the student the vast range
of social situations within a typical work environment. However, the students can
be directed to similar recording techniques, as described in Chapter 11.
In particular, this exercise is a good chance to introduce the use of video or
audio-recording equipment if it is available. The students will soon learn how
difficult it is to position a microphone so that it picks up more than one worker and
so that it does not get drowned by the sound of a typist. Similarly, they can gain
first hand experience of the problems of static camera positioning. If you are lucky,
you may have access to multiple cameras and split screen recording equipment.
Although such equipment can be useful, its lack is not a disaster. Many ethnographic studies use only a pencil and paper, with perhaps some audio recording.
However, developing appropriate shorthand then becomes essential (see Suchmans book for examples of appropriate notation).
Some forms of communication in the office can be quite subtle, perhaps overhearing of conversations, or noticing when something is being written on a wall
calendar. Developing an observing eye for such subtleties is largely a matter of
practice.

June 4, 1993 17:34

exercises

Sheet number 93 Page number 91

Chapter 15

Multi-sensory Systems

Exercise 15.1
What factors are likely to delay the widespread use of video in interfaces?
What applications could benefit most from its use?

Answer
Some of the factors are the costs in terms of hardware and software for compression
and decompression; the slow speed due to the high bandwidth; the overall cost
of equipment (for example, camera, video, CD); the lack of design tools to exploit
video; the lack of specialist skills amongst designers. Many applications have
been suggested as candidates for the integration of video. Educational systems,
games and help systems are liable to benefit since information can be passed
more clearly and memorably and new dimensions added. Other areas such as
virtual reality can use video together with graphics in the creation of their artificial
worlds. CSCW systems can use video to provide a face to face communication link
between distributed workers (see Chapters 13 and 14). However, although these
appear to be areas where video has a promising future, its use needs to be carefully
considered and its consequences investigated. It may be that it will not fulfill its
initial promise.

Exercise 15.2
In Section 15.11.1, we look at alternative interfaces for users with special
needs. How could standard interface design be improved to provide better
access to these users?
91

June 4, 1993 17:34

92

exercises

Sheet number 94 Page number 92

Multi-sensory Systems

Answer
This depends on the particular needs of the user but in general the more an interface
relies on a single channel for information (such as the visual channel) the more
likely it is to be unusable by some people. Multi-modality and redundancy in the
interface will alleviate this problem providing alternative views of the information
and so not excluding a user group with restricted use of the primary channel.
It is more satisfactory to encourage multi-modality as a design principle than to
provide special systems or add-onsu`u` to cater for extra-ordinary user groups. This
can also benefit the average user since it reinforces the information over several
channels. However this may not be possible in all cases: clearly, for example,
a user who has limited control of hands and voice may need special equipment
such as the Eyegaze system which will track eye movements. Where possible such
specialized equipment should be integrated with standard applications to enable
the disabled user to use readily available machines and software in common with
other users. This is particularly important in the work environment.

Exercise 15.3
Think of a set of naturally occurring sounds to represent the operations
in a standard drawing package (for example, draw, move, copy, delete,
rotate).

Answer
This can exercise the imagination! Are there natural analogies? For example does
the physical action say of drawing have a sound associate with it? The sound of a
pencil on paper may be appropriate but is it identifiable? Similarly a photocopier
whirring could represent the copy operation, and tearing paper delete. Rotate and
move are more difficult since the physical operation is not associated with a sound.
Perhaps direction and movement can be indicated by sounds becoming nearer or
more distant?

Exercise 15.4
Experiment with Hypercard or another hypertext system if you have access
to one. As you work through the system, draw a map of the links and
connections. Is is clear where you are and where you can get to at any
point? If not, how could this be improved?

June 4, 1993 17:34

exercises

Sheet number 95 Page number 93

Multi-sensory Systems

Answer
This is an experimental exercise which requires access to a hypertext system. It
can be used as the basis for a practical class, in which the students analyze the
effectiveness of the system. Drawing he map has two purposes: one is to reinforce
the overall structure of the hypertext; the other is to test the navigational support
that is available. Whether it is sufficient will depend on the system under scrutiny,
but possible improvements would be to provide an explicit map, escape buttons,
explicit paths to core material. The system may of course incorporate such features.

Exercise 15.5
Data visualization techniques have often increased our comprehension
of phenomena: consider the effect that 3-D graphics has had on looking
at complex models such as those of the atmosphere or the ocean, or in
understanding the structure of molecules. What do you consider to be the
areas that may benefit most from virtual reality visualization techniques?

Answer
This could be expanded to allow the student to carry out a brief literature review on
the applications of virtual reality techniques. In general they are most promising in
areas which cannot be explored in reality, either because it is physically impossible
(such as in the manipulation of molecules discussed in Section 15.11.2) or physically
dangerous, such as in space. Models can be physically manipulated and interacted
with using virtual reality methods rather than simply observed. Chemical and
medical research, meteorology, oceanography, seismology are all areas which could
benefit from such techniques.

Exercise 15.6
Virtual reality has found a number of applications in the games market. Is
this a suitable use of such technology? Discuss the possible benefits and
disadvantages of exploiting leading edge technology in a leisure market.

Answer
The leisure market is important to the computer industry and, being commercially
lucrative, can provide an impetus for development which can then be exploited in
other spheres. If virtual reality takes off in games it will result in the development of

93

June 4, 1993 17:34

94

exercises

Sheet number 96 Page number 94

Multi-sensory Systems
cheaper equipment and more robust techniques, since the games customer tends
to be young (and not overly rich) but with sophisticated expectations. Indeed,
many research projects in universities are currently using equipment developed
specifically for games, since the more powerful, general purpose equipment is prohibitively expensive. Another possible advantage is that it makes the technology
familiar and therefore accessible and acceptable.
A possible disadvantage is that the use of this technology in games may trivialise it so that it is not considered as a serious solution to other more weighty
problems. However, this seems less likely than the advantages outlined above.

Exercise 15.7
What are the major achievements of computer vision? What are the major
problems that remain to be solved?

Answer
Computer vision is being used in a number of application areas, where the images
are used to provide input to the computer. For example, face recognition in
security systems, monitoring airports and public areas for suspicious behaviour,
analyzing scans and xrays, particularly to identify objects in packages, analyzing
aerial photographs for cartography and military uses. The success of such systems
has been limited and most computer vision is still at the research or experimental
stage.
There are a number of problems still to overcome. These include coping with
variation in images, such as orientation, shading and light changes, dealing with
scenes which include partially occluded objects, speed of scene analysis.
Again this exercise provides the student with a focus fro a brief literature
review on the subject. A textbook on computer vision or artificial intelligence is a
good place to start.

June 4, 1993 17:34

exercises

Sheet number 97 Page number 95

Design Projects

15.1
15.1.1

Icon Design
Aim and structure

This is an exercise in design which should be attempted towards the beginning of

the course. It fits in well after the introductory part of the book has been covered
(Chapters 1-3). The aim is to encourage students to think about some of the
problems involved in providing clear, unambiguous interfaces, even to familiar
applications. It also helps the student to begin to evaluate designs critically. They
should be able to draw on their knowledge of human cognition and of interaction
styles but do not need to use any particular modelling or evaluation techniques.
These will come later.
The exercise should be attempted in groups. There are two parts to it. First,
each group should discuss the problem and try to reach a consensus on their
designs. They can choose to draw upon their experience of other systems if they
wish, although it should be pointed out that the developers of the new package do
not wish to become engaged in law suits over copied designs! They should then
sketch their designs as clearly as possible. It is easier if graph paper is available
for this part.
Secondly, the group should display their designs and assess them and those of
the other groups. This should be done critically but fairly it is worth pointing
out to them that their work is being assessed as well! The students should identify
what they believe to be good designs and most importantly the reasons why they
are good.

15.1.2

Instructions to the student

You are part of a development team charged with producing a new drawing
package. You have to develop the full set of icons. This is a new product and so
there is no set house style. It is to appear on a new platform and so does not have
to follow any set of user interface style guidelines. The platform usually supports
95

June 4, 1993 17:34

96

exercises

Sheet number 98 Page number 96

Design Projects
colour on a high-resolution bit-mapped monitor, but icons must be suitable for use
on black & white as well as colour screens design in black and white first then
add colour. Text can be used if required. Each icon needs an unselected and a
selected state.
The following is a list of all the functionality required: you may choose a form
other than an icon to provide the function but you should justify your choice.

















freehand draw with pencil

freehand draw with paintbrush
smoothed freehand drawing
draw rectangular box
draw round-edged box
draw ellipse
fill or shade items
move or resize item
write text
draw irregular polygon
draw straight lines
draw arc
scale object
rotate object
edit shape

Icons should be designed on 64x64 grids. Try to draw them as clearly as

possible and use colour if required.
When you have completed your designs, display them, marking them clearly
with their associated functionality. Now look the designs produced by other
groups. Try to assess them using the following questions to help you:






is the meaning of the icon clear?

is colour used sensibly and consistently?
will the icons meaning still be clear in black and white?
what is the overall effect of the icons when viewed together?

June 4, 1993 17:34

exercises

Sheet number 99 Page number 97

VCR Remote Control

how original are the designs and is the originality effective or simply novelty
for noveltys sake?

Discuss your assessment within your group and with other groups. How does
your work compare to others? Make any adjustments you think necessary to your
designs.

15.2
15.2.1

VCR Remote Control

Aim and structure

This exercise requires the student to design an interface for the remote control for a
VCR. This involves first determining the tasks that the remote control must support
(these may be a subset of those that the VCR supports) and second determining an
appropriate dialogue design.
Students should be encouraged to work in groups and to utilize the modelling techniques for task analysis and dialogue design that they have studied (see
Chapters 7 and 8). By the end of the exercise they should be satisfied that their
design supports the tasks identified. It is assumed that most students will be familiar with VCRs and the tasks that users wish to peform with them. It may be helpful
to show them various examples of VCR remote controls to get them started.

15.2.2

Instructions to Students

You are asked to design a remote control for a VCR which will support the most
common tasks required by video users effectively. Your first job is to identify these
tasks and the subtasks that are required to perform them. Use a task analysis
method (Chapter 7) to produce a detailed description of these tasks.
Having identified the key tasks, you now have to design an interface to support
them. Use a dialogue notation (Chapter 8) to describe the sequence of actions
associated with each task and show how these relate to each other. When you are
satisfied with your dialogue, produce a mock-up of the interface showing how the
user will perform different tasks. Ensure that your design is consistent with your
dialogue description.

97

June 4, 1993 17:34

98

exercises

Sheet number 100 Page number 98

Design Projects

15.3
15.3.1

Mobile phone
Aim and structure

This is a design exercise in two parts. The first part uses information from users
to develop a design for a mobile phone. The second part uses different evaluation
techniques to assess the design. There are two aims: to demonstrate how the
views of users can be incorporated into design and to demonstrate the different
information that can be gained by using different evaluation methods.
As with the previous project, students should be encouraged to use any modelling technique that will help them in their design. Again they should work in
groups or pairs.

15.3.2

Instructions to Students

You have been asked to design a mobile phone taking into account the user views
expressed below, determined from market research studies. There are a number of
factors that have to be considered in the design, ranging from the way information
is entered into the system and the manner in which is presented back to the user,
through aesthetic judgements to the functionality that the system should offer.

15.3.2.1

Design

Read the user responses given below and then analyze them under the following
headings:
Functionality: You need to determine the tasks which users wish to perform
with the phone, and how they want to go about performing those tasks. You need
to decide what features your phone should offer and what are dispensible. You
are constrained by real-life costs and so will have to trade off size for weight and
battery life, cost against functionality, and so on.
Try to identify the real issues that people consider when using a mobile phone.
Try to see what factors will influence their decision about which phone to buy. It
is these that must motivate your design. Make a list of the features that you feel
your phone must have.
Interface: How are you going to present all this information to the user? What
style of output is most suitable? What is the best way to interact with it? Will you
present all the information at once? What input characteristics will your phone
have: how will numbers be entered, functions accessed, and so on?
Having thought about these things, consider that you are also trying to keep
the electronics inside the phone as simple as possible. Remember too that people

June 4, 1993 17:34

exercises

Sheet number 101 Page number 99

Mobile phone
will not carry a manual for their phone around with them, and so they will need
to be able to understand what to do, or to be reminded by the interface itself.
Aesthetics: Consider issues of overall colour, size, shape, texture, of button
spacing, layout, colour and feedback, display size and placement, information
display.
You will use this analysis as the basis of your design. Modelling techniques
such as task analysis can help to clarify this which will in turn make the process of
design easier.
Having decided upon all the relevant information, now design your phone.
You have to produce two views of the phone in all. The first is a picture of the
phone as it appears to the user, with no additional labelling - make this as high
quality as you can. Imagine it is the sort of picture that would appear in a brochure
advertising the phone. The second view is of the same phone, but should be
viewed from above. An additional set of pictures of parts of the phone should
also be provided that should enable you to run through a typical interaction with
a prospective user, showing the display state when they perform certain actions to
achieve a certain function.
Either do all of these designs separately on paper, using colour, shading and
different viewpoints as appropriate, or utilise OHPs to overlay the basic design.
This is particularly useful for the second view of the phone as it allows you to
update the display state without requiring redrawing the rest of the phone.
Again modelling techniques can help in the development of an effective dialogue. Consider for example using a dialogue notation to describe the details of
your interface or performing a GOMS analysis of the main tasks supported by your
design.

15.3.2.2

Evaluation

The second part of this exercise concentrates on the evaluation of your design and
looks at two different evaluation methods: heuristic evaluation and cooperative
evaluation.
The first evaluation method is heuristic evaluation (see Section 11.4.2) You
will evaluate the design against a set of general usability criteria laid out below:







system behaviour is predictable

system behaviour is consistent
feedback is provided
the users memory is not overloaded
dialogue is task-oriented

99

June 4, 1993 17:34

100

exercises

Sheet number 102 Page number 100

Design Projects
These criteria are discussed in more detail in Section 11.4.2. The aim of the heuristic
evaluation is to debug the design; to highlight the points in the design that are
inconsistent or likely to cause users problems.
Exchange designs with a colleague. Try to evaluate their design using all of
their design material and the usability criteria given above. During your evaluation, consider if there are other general criteria that are important considerations
for this particular example; if so, include them. You are aiming to see where the
design is successful and where it fails; it may assume too much, or too little, it may
not offer any feedback, or present an unclear view of what is happenning. The user
may be expected to remember too many things at once, or there may be different
ways to do similar things. Make notes of your evaluation.
Return the evaluation and your notes on it to the original designer, and repeat
the exercise on a different design.
At the end of this stage, you should have done two heuristic evaluations, and
have received two independant evaluations of your design. Go through these
evaluations, collating similar problems and resolving differences between reports,
until you have an accurate summary of the evaluation.
Cooperative evaluation is another evaluation method, in which the user is
encouraged to see himself as a collaborator in the evaluation rather than just a
subject (see Section 11.5.2). In pairs, take it in turns to be an evaluator and a user
for each design.
As an evaluator spend a few minutes thinking of some scenarios and tasks for
the user to perform. For example, to look up a telephone number, to meter a call,
or to dial a number. Include some complex tasks as well as some simple ones. Ask
the user to step through these tasks using each design. Take notes on the users
actions, comments and any problems. The fuller the detail in the protocol, the
better.
As a user, follow the evaluators instructions. Try to give as much information
as possible. Explain what you are doing at all stages and why. Answer the
evaluators questions as clearly as possible.
Look at the results from the heuristic and cooperative evaluations. Notice
where they highlight problems in the design. Also look for instances in which the
designallows tasks to be easily accomplished. Compare the two techniques and
note anything which is identified by one technique and not the other.

15.3.2.3

Redesign

Reappraise your design carefully in the light of the evaluation process. Carefully
look at things that cause confusion, at the functions that are there but may never
be used, and at the comments made about the good and bad points of your design.
Using your experience of others designs and the comments on your own, you have
a great deal of information about what should be included, and excluded, from
a new design. It may help to work in groups of four or five, for this final stage,
though it is up to you. Pool your designs and try to come up with a new version of

June 4, 1993 17:34

exercises

Sheet number 103 Page number 101

Mobile phone
the phone. It should incorporate all the good features of the previous prototypes,
whilst avoiding all the problems that they may have suffered. Once decided upon,
each of you should produce a final labelled drawing of the ideal phone.

15.3.3

Mobile Phone: User Responses

A group of people were interviewed to obtain their views on the features they
would expect to find in their ideal mobile phone. Some of the people currently
use a mobile phone, others do not. Given below are the (edited) texts of their
responses.
A: I want to keep track of the number of calls that I make, and the cost of the
calls. Since I use the phone for business it has to be small but powerful but cant
cost a lot to buy. I dont want to fuss with that stupid airel thing that you have to
flip up in order to use it and I want to be able to keep ringing people back if theyre
busy - as I said it has to be hand-held but thats it really.
B: Ohhh - a mobile phone! Well, it just has to ring people and let people ring
you, doesnt it.
C: It must be able to store telephone numbers in as I wont have my telephone
book with me but I dunno how Id search for the numbers - perhaps a pad on
which you could write them down would be a good idea - no thats daft really isnt
it I mean its supposed to be electronic I suppose what about a volume control for
ringing. I remember someones phone going off in a meeting it was really loud
and she was so embarassed .....what about the buttons will it have press buttons
or a dial? Oh I can decide well I like the dial personally I think its aesthetically
pleasing but some people like buttons I dunno really. Running them costs a lot of
money so I suppose Id like to be able to tell how much Id spent that day on calls
sort of get my bill as I go along I suppose. Will I be able to turn it off? Oh - I have
to decide do I? Well I suppose I should have it with me and on all the time but
what if I didnt want to be interrupted so I suppose Id want to turn it off ..... but
then I may miss important calls - can I have an answering machine built in as well
or something so that it answers my calls if Im not there - that would be nice.
D: I wouldnt use one ever - yuppie toy!
E: Well, the buttons would have to feel nice as I hate those plasticy tacky
ones yu get on those 8-99 phones. Its difficult to tell when the button has been
pressed sometimes. Some sort of display would be good - I suppose it should
have information on the status of things - battery maybe and the signal strength,
th number Im calling, the previous numbers so that I can scroll back and get them,
and perhaps an address book thingy that has names and addresses and telephone
numbers in. Hmmm thats a bit bulky right - Id want it to be pocket sized to. Oh
and the batteries should have to last for ages and ages and I want to be able to
charge it up anywhere like in the car and so on.

101

June 4, 1993 17:34

102

exercises

Sheet number 104 Page number 102

Design Projects
F: Oh - Id never have one because what if I lost it? Its not the cost of the
phone thats the problem its that someone may ring up their friend in Australia
and Id have to foot the bill. It would be okay if only I could use it.
G: Wow - mobile phones are great - I want one that rings people up, takes
messages, will allow me to send faxes and receive them, will act as a calculator
as well - I mean theres all those number buttons already right? - and an address
book and small word processor - its got to have a display and things right? so it
could also be a personal organiser and interface to PCs and Suns and Macs and
things and it must be small; thats for certain, it has to be small and the battery has
to last for a week or so without being recharged and handheld - no I said that errrm thats it. Oh and I want to use it with the interface bit so as I can read my
email on the Suns when Im away and stuff.
H: Easy: 1. see the number Im ringing. 2. be able to redial easily when a
number is busy. 3. see the battery strength. 4. see how much its cost me. 5. have
nice buttons to press with good feedback. 6. not have to reenter all a long number
if I make a mistake. Thats all
I: Id like a system that has an address book in it, but it doesnt have to be too
powerful. Then I want to be able to see what number Im ringing and re-ring it
if necessary and see how much the calls have cost. A good aerial will be needed
because good reception is really important. A way of barring calls international
numbers would be good and a way of barring others from using it would also be
nice. Id want to turn it off occasionally but I dunno what would happen if anyone
tried to ring me.
J: I like them, especially the grey ones - theyre better than the black - and the
pretty light that comes on thats good too dunno what its for though.
K: I dont like them - because Im a bit deaf theyre too quiet for me and I cant
hear what people are saying.
L: Ha! what would be nice is a phone with not too many buttons! They all
have too many buttons nowadays and I hate that. Keep it simple, I reckon.
M: I dont understand them. I never know which button to press to get the
phone to ring and which to press to stop a call, and I can never work out how to
answer it either - do you just pick it up or what? And what are the letters on the
number buttons for? You know the 1 has abc on it and so on?
N: Whats a mobile phone - a cordless one? Oh, one of those things. Naw,
I dont want one of those - theyre like half a housebrick.....oh, I can choose can I
- well I just want the same as Ive got at home - small, you know, but having it
portable. Thats all I want.