Lawson B. How Designers Think: The Design Process Demystified
Подождите немного. Документ загружается.
robustness. Although it may prove possible to measure designs
on crude scales of satisfaction for each of these factors, they remain
difficult to relate. Thus a very lightweight lawnmower while being
easy to manoeuvre and push might also prove to be noisy and easily
damaged. For such an item there is no one right answer since differ-
ent purchasers are likely to place different values on factors such as
manoeuvrability or reliability. The sensible manufacturer of such
equipment will produce a whole range of alternative designs each
offering different advantages and disadvantages. The problem of rela-
tive values, however, becomes much more critical when design deci-
sions are being taken for large numbers of people who may not have
the choice available to the purchasers of new lawnmowers. Examples
of such design problems include public sector housing or a new
school, the routeing of new roads or the siting of factories. Inherently,
such projects involve varying degrees of benefit to some and losses
to others. A new motorway may well save a long-distance motorist’s
time and relieve congestion in nearby towns but, unfortunately, it
may also subject local residents to unwanted noise and pollution.
The attraction of a common metric
An attractive way out of all the difficulties we have seen in this chapter
would be if we could reduce all the criteria involved in design to
some common scale of measurement. Cost-benefit analysis relies
upon expressing all factors in terms of their monetary value, thus
establishing a common metric. Attempts have been made to apply
cost-benefit analysis techniques to the kinds of design problems
where there are both gainers and losers. Unfortunately, some fac-
tors are rather more easily costed than others. This is perhaps best
illustrated by reference to one of the most well-known applications
of cost-benefit analysis, the Roskill Commission on the siting of the
third London airport. After a number of preliminary stages during
which some seventy-eight sites were considered, the commission
narrowed the choice down to four sites at Cublington, Foulness,
Nuthampstead and Thurleigh which were then compared using
cost-benefit analysis. Even the grossly simplified diagram repro-
duced here gives some idea of the complex array of effects which
the various interested parties could be expected to have on each
other as a result of such a project (Fig. 5.6). In fact there are many
other much wider effects not shown which include such matters
as the distortion of the national transportation network resulting
HOW DESIGNERS THINK
78
H6077-Ch05 9/7/05 12:27 PM Page 78
from the provision of new forms of access to the chosen site. For
example, the opening of an airport at Cublington would have
resulted in the closure of the existing Luton airport which would
have been too close for air traffic control procedures.
Many of the benefits of the airport in terms of the profits to the
various transportation authorities and other companies were rea-
sonably easy to calculate for each site and could be set against the
profits lost from the existing use of land. The costs of providing the
access transportation to each site and the costs in terms of journey
time were also fed into the equation. Losses in terms of reduced
amenity, however, proved more difficult to assess in purely mon-
etary terms. These effects range from otherwise unwanted expend-
iture resulting from people having to leave their homes, through
such factors as the depreciation in value of property in the sur-
rounding area to the noise annoyance caused by the operation of
the airport.
Such a public use of cost-benefit analysis revealed many of
the real dangers involved in basing decisions on the quantification of
qualitative factors such as the amenity of an environment. Obviously
the success of such a process is contingent upon the assumption
that all the costs of amenity loss have been correctly valued. The real
difficulty here is that such valuations are unlikely to be arrived at by
consensus in a pluralistic society. This was demonstrated when the
RIBA publicly expressed its concern at the valuations placed on both
gains and losses and pointed out the many minor losses not costed
which might have a large effect cumulatively:
MEASUREMENT, CRITERIA AND JUDGEMENT IN DESIGN
79
builders
buildings
land
labour
supplies
employment
transport
supplies
buildings
equipment
floor
space
airport
authority
landowners
firms at
airport
passengers
airlines
residents
local
business
landing
services
goods
noise
Figure 5.6
A simplified diagram of the
interactions between the parties
involved in a new airport
H6077-Ch05 9/7/05 12:27 PM Page 79
An hour lost by an air traveller is valued very generously, taking into
account business overheads as well as salary, but an hour of sleep lost by
those living outside the area of major impact is given no value whatever.
(RIBA 1970)
The costing of noise annoyance or the value of quiet had proved
difficult enough for the Roskill Commission, but when consider-
ations of the conservation of wildlife at Foulness were introduced
to the argument the whole decision-making process began to split
at the seams. Cost-benefit analysis was clearly incapable of develop-
ing one equation to balance the profits of an airport against
the loss of a totally unproductive but irreplaceable and, some
would say, priceless sanctuary for birdlife. The Roskill report itself
recognised the futility of attempting totally objective judgement in
comparing the Cublington and Foulness sites. The choice was
between the damage to the value of Aylesbury and the loss of a
fine Norman church at Stewkley or the ruining of the Essex coast-
line and probable extinction of the dark-bellied Brent goose:
As with much else in this inquiry there is no single right answer however
much each individual may believe there is. For us to claim to judge
absolutely between these views (the importance of conservation of
buildings or wildlife) is to claim gifts of wisdom and prophecy which no
man can possess. All we can do is respect both points of view.
(Roskill Commission Report)
Even the costings of the more ostensibly easily quantifiable factors
proved extremely debatable. For example the cost-benefit research
team itself revised the assumptions on which total construction costs
had been based. This change proved so drastic that Cublington
moved from being the most costly to the least costly of the possible
sites in this respect. The inquiry proceeded until it gradually became
apparent that many of the fundamental underlying assumptions nec-
essary for the cost-benefit analysis could similarly be challenged.
The indecision which resulted at least in part from the discrediting of
the technique led to many years of procrastination before an airport
was finally built at Stanstead. Perhaps the last word here should
come from Professor Buchanan, a member of the Commission who
became so concerned that he published a minority report:
I became more and more anxious lest I be trapped in a process which
I did not fully understand and ultimately led without choice to a conclu-
sion which I would know in my heart of hearts I did not agree with.
Recently there has rightly been more emphasis placed on the
ecological implications of design decisions. Most of the energy
HOW DESIGNERS THINK
80
H6077-Ch05 9/7/05 12:27 PM Page 80
consumed in the developed countries is connected with the manu-
facturing and use of products. A very high proportion indeed is con-
nected with the construction industry. Similarly, levels of pollution
and atmospheric emissions are heavily influenced by the decisions of
industrial designers, architects and town planners. All this leads us to
want more information on the true impact of design decisions, not
just at the stage of constructing but in terms of the full life cycle.
Again legislation is increasingly setting, and then changing, limits on
energy consumption and pollution. Most designers are probably
very conscious of the need to improve our world in this way, but find
it extremely difficult to incorporate findings and recommendations
into their design process. The findings and data are seldom clearly
expressed in a form which a designer can make sense of. Just as it is
increasingly difficult to know what it is safe and healthy to eat, so
designing in an ecologically sound way is surrounded by myths,
campaigns and, sometimes, deliberately misleading data. In all this
confusion, however, designers cannot usually procrastinate as did
those deciding on the third London airport. They simply must get on
and make the decision in as integrated and sensible a way as they
can. Their decisions then remain very visible and easy to criticise as
data becomes more clearly available!
Objective and subjective decisions
In the final analysis it seems unreasonable for designers to expect
to find a process which will protect them from the painful and diffi-
cult business of exercising subjective judgement in situations
where both quantitative and qualitative factors must be taken into
account. The attempt to reduce all factors to a common quanti-
tative measure such as monetary value frequently serves only to
shift the problem to one of valuation. The Roskill Commission on
the siting of the third London airport provided one further lesson
of importance here. Designers and those who make design-like
decisions which profoundly affect the lives of many people can no
longer expect their value judgements to be made in private. Such
large-scale design processes must clearly invite the participation of
all those who will be substantially affected. However, we must
not expect the design process to be as clear, logical and open a
process as the scientific method. Design is a messy kind of busi-
ness that involves making value judgements between alternatives
that may each offer some advantages and disadvantages. There is
MEASUREMENT, CRITERIA AND JUDGEMENT IN DESIGN
81
H6077-Ch05 9/7/05 12:27 PM Page 81
unlikely to be a correct or even optimal answer in the design
process, and we are not all likely to agree about the relative merits
of the alternative solutions.
References
Alexander, C. (1964). Notes on the synthesis of form. New York, McGraw Hill.
Alexander, C. (1966). ‘A city is not a tree.’ Design 206: 44–55.
Archer, L. B. (1969). ‘The structure of the design process.’ Design Methods
in Architecture. London, Lund Humphries.
Boje, A. (1971). Open-plan Offices. Business Books.
Broadbent, G. (1973). Design in Architecture. New York, John Wiley.
Chermayeff, S. and Alexander, C. (1963). Community and Privacy.
Harmondsworth, Penguin.
Lawson, B. R. (1975a). Heuristic science for students of architecture.
University of Sheffield Department of Architecture.
Lawson, B. R. (1975b). ‘Upside down and back to front: architects and the
building laws.’ RIBA Journal 82(4).
Lawson, B. R. (1979). ‘The act of designing.’ Design methods and Theories
13(1).
Lawson, B. R. (1982). Science, legislation and architecture. Changing
Design. New York, John Wiley.
RIBA (1970). ‘The third London airport: choice cannot be on cost alone
says RIBA.’ RIBA Journal 77(5): 224–225.
Savidge, R. (1978). ‘Revise the regs: the plan revealed.’ The Architects’
Journal 167(14).
Stevens, S. S., Ed. (1951). Handbook of Experimental Psychology. New York,
John Wiley.
HOW DESIGNERS THINK
82
H6077-Ch05 9/7/05 12:27 PM Page 82
6
A model of
design problems
As an artist I did not set out to make the public understand but to
find problems for myself of space and form, and to explore them.
Henry Moore (on his 80th birthday)
There is nothing absolute about deriving architecture from functional
manipulations. When you look at a plan by Corb, these wonderful
shapes that he has rationalised out of bathrooms and things, I mean
they are magic really, they are completely magical diagrams but
I would much rather have the magic without the spurious functional-
ity in the way.
John Outram
Analysing design problems
In this chapter we try to analyse the structure of design problems.
As with the rest of the book, this analysis is primarily directed at
those problems which are solved by three-dimensional design, but
may in many cases be sufficiently generic to apply at least in part
to graphic design and some kinds of engineering. This analysis will
be based on an investigation of the generators of design problems,
their domain of concern and their function. From this study we
shall be able to assemble the building blocks which make up a
model enabling us to understand the nature of design problems in
all their variations. This model has been found useful over many
years of teaching and researching the design process. It is pre-
sented here in order that we may further understand what makes
design problems so special and thus gain some insights into how
designers think and why.
H6077-Ch06 9/7/05 12:27 PM Page 83
The generators of design problems
At first sight it may seem obvious where design problems come
from. Clients bring them to designers! As we shall see, whilst that
is often true it is not always so, and it turns out to be only a small
part of the story. It is certainly possible for designers to discover
a problem without a client and much interesting design work is
done under exactly these clientless conditions. We also to need to
draw a careful distinction between the clients who present prob-
lems to designers and the ultimate users of the outcome. As we
shall see, clients may or may not be the users of design. We have
seen in the last chapter, how legislators can often pose consid-
erable problems for the designer and they may sometimes even
be in conflict with the client. Town planning legislation for exam-
ple exists chiefly to protect the general public from the possible
selfish excesses of individual architectural clients. Whether such
development control is actually so beneficial, however, is probably
debatable! However, we are in danger of getting ahead of the
argument.
Clients
In design, the problem usually originates not in the designer’s
mind but with a client; someone in need who is unable to solve
the problem, or perhaps, even fully to understand it without help.
Whilst the fortunate artist may occasionally be commissioned, the
designer almost always works this way. The design task, albeit
ill-defined, is usually initially generated and expressed by a client.
However, it is quite misleading to think that clients are a homo-
genous group. In many commercial situations the client may be
represented by a professional, acting in that capacity more or
less as a job. At the other end of the scale, many buildings
are commissioned by people who have never acted as a client
before. Sometimes the designer will work with an individual
client, and at other times the client body may be represented
by a whole committee. In the case of very large buildings com-
missioned by institutions or companies the programme may last
several years and the membership of the client committee may
change substantially.
The architects Stirling and Wilford have had considerable experi-
ence of these large institutional clients and have built many civic
HOW DESIGNERS THINK
84
H6077-Ch06 9/7/05 12:27 PM Page 84
and educational buildings. Michael Wilford has emphasised the
importance of the role of the client in the design process:
Behind every building of distinction is an equally distinctive client, not
necessarily high profile, but one who takes the time and trouble to
comprehend the ideas of the architect, is supportive and enthusiastic,
who is bold, willing to take risks and above all can hold his or her nerve
during the inevitable crises.
(Lawson 1994b)
This suggests quite firmly that Michael Wilford does not just see
the client as the source of the brief but as a creative partner in the
process. The architect Eva Jiricna agrees with this by suggesting
‘the worst client is the person who tells you to just get on with it and
give me the final product’ (Lawson 1994b). The client then is the
most obvious example of a source of design problems and con-
straints. Ideally, and frequently those constraints can be explored
creatively through an interaction between designer and client. It is
certainly misleading to think that a client simply presents a designer
with a complete brief in which the problem is totally defined and the
constraints clearly articulated. The relationship between client and
designer itself actually constitutes a significant part of the design
process. The way that designers perceive and understand problems
is to some extent a function of the way this relationship works.
Users
A great deal of design today is commissioned by clients who are
not themselves the users. Public architecture such as hospitals,
schools or housing is usually designed by architects who have
relatively little contact with the users of their buildings. Industrial
design and graphic design are directed at a mass market and are
usually commissioned by commercial clients. The traditional image
of the designer establishing a personal relationship with a client/
user is grossly misleading. Even architects commissioned to design
new buildings for large organisations such as universities are likely
to be buffered from the actual users by a client committee or even
a full-time buildings department. Frequently communication
between designers and their users is both indirect and, as John
Page has argued, filtered by organisational politics. In his study
of ‘planning and protest’ (Page 1972) he describes the ‘people
barriers’ erected in many organisations to prevent too much dis-
ruptive user feedback reaching designers.
A MODEL OF DESIGN PROBLEMS
85
H6077-Ch06 9/7/05 12:27 PM Page 85
In local authorities, for example, both the politicians and admin-
istrators may attempt to establish themselves as the communica-
tion channel between the designers and the users outside in order
to force through policy or maintain a powerful position in the sys-
tem. On balance such organisational barriers, whatever advantages
they give to the client body in terms of increased control over
the designer, serve only to make the designer’s task of understand-
ing the problem more difficult. Even if there are not barriers there
are what Zeisel (1984) has called ‘gaps’. He referred to ‘paying
clients’ and ‘user clients’. He showed that while there might
often be good communications between designers and paying
clients, both have a gap in their communications with user clients
(Fig. 6.1). In a more recent piece of empirical work Cairns (1996)
not only demonstrated the existence of these gaps in architectural
design, but also that neither architects nor their clients were always
aware of these gaps.
As many young designers must have found on leaving design
schools, it is one thing to design for yourself but quite another to
design for a real client with personal and institutional prejudices
and biases. When that client is not even the prospective user of the
design, the problem becomes even more remote. This increasing
remoteness of designers from those for whom they design has cre-
ated the need for user requirement studies. Almost in desperation
designers have turned to social and human scientists from ergono-
mists through architectural psychologists to urban sociologists to
tell them what their users actually need. By and large this liaison
between design and social science has not been as practically
useful as was first hoped. Social science remains largely descriptive
while design is necessarily prescriptive, so the psychologists
and sociologists have gone on researching and the designers
designing, and they are yet to re-educate each other into more
genuinely collaborative roles. Meanwhile the communication
HOW DESIGNERS THINK
86
designers
paying
clients
users
gap
gap
Figure 6.1
Zeisel’s user-needs gap model
H6077-Ch06 9/7/05 12:27 PM Page 86
between the creators and users of environments often remains
uncomfortably remote.
So users are generally more remote from designers than clients.
Whilst the designer may be able to interact with a sympathetic
and motivated client, there may be no formal access to users
at all.
Designers
It is sometimes difficult to separate design from art. The products of
design are frequently seen by the public as artistic, even sometimes
actually as ‘works of art’, and designers themselves are indeed also
often artists. Even the drawings generated by designers to illustrate
their schemes can sometimes easily be confused with works of art.
Whether or not an object can rightly be described as a ‘work of art’
is a matter which lies beyond the scope of this book. What is of
importance here is not the product but the process. The creative
process which may give rise to a work of art undoubtedly shares
much in common with the design process, and many of the same
talents may be needed for both. Designers, like artists, are
expected not just to solve problems but to bring their issues and
concerns into the process too. In this sense, however, the designer
is usually rather more constrained than the artist. The artist may
respond to the work in progress and is free to shift attention and
explore new problems and territory. Such artistic issues are rarely
clearly articulated by the artist beyond the work. It is usually critics
and historians who retrospectively interpret and identify the issues
which appear to them to have been uppermost in the artist’s
mind. When asked by a music critic to explain one of his operas,
Wagner is reported to have responded rather testily ‘but it is the
explanation’.
The designer is usually expected to contribute problems too.
In this sense designers are assumed by their clients to be artistic
and their role to be at least partly interpretative. An architect’s
client expects rather more than just a house with rooms of
appropriate sizes and relationships. The clear expectation is that
an architect will consider issues of, for example, form, space and
light, and through this create not just a building but what we
call architecture. This client–designer relationship works both
ways, for while the designer expects to be given some freedom
in the definition of the design problem. It is also quite likely that
A MODEL OF DESIGN PROBLEMS
87
H6077-Ch06 9/7/05 12:27 PM Page 87