Abstract
Much traditional HCI research has concentrated on
routine, well-defined and stable, tasks or low level
computer support for complex tasks, e.g. spell-checking
for someone writing a book. Increasingly, however,
interest is moving to the support of people involved in
creative tasks. This is the topic of the panel. Design
and the visual arts will be used as typical examples of
creative work and visions of computer futures and their
cultural and social implications are explored.
Keywords:
creativity, interaction, design, art,
emergence, distribution, concurrency.
Introduction
Computer systems for professional workers are often
constructed in relation to a problem solving paradigm.
On the other hand, studies of professional people at
work suggest that they spend much more time in
problem formulation than they do in problem solving.
Problem solving requires expertise but problem finding
requires creative thought. Should we aspire just to
automate expertise or should we aim to amplify human
creativity?
Just what is stimulating to creative thought and what is
inhibiting? For example, a software critic can be very
helpful in bringing errors to the attention of the user,
but the critics knowledge can include conventional
wisdom. A creative act may often involve
contradicting a standard convention. So does the critic
help or hinder? What should the research agenda be for
advancing support for creative thought and action?
POSITION STATEMENTS
Ernest Edmonds
For computers to support human creative thinking, they
must be able to keep up with human recognition of
emergent ideas. This implies that the system is not
based upon a well ordered object set but has pattern
recognition capability that can find the new objects, as
they emerge, with minimum human guidance.
Emergence is fundamental to creative thought in the
sense that we find it hard to qualify an idea as creative if
it is clearly implied by the preceding conditions. The
creative thought introduces something new. In studies
of design, for example, we see the reshaping of the
significant creative events. In a recent study of
innovative bicycle design it was shown that the designer,
Mike Burrows, shifted his thinking from the
conventional tubular frame to the concept of a single
"monocoque" whole that could not have been inferred
from the earlier model. As he considered smaller
tubular frames he came to see the possibility of filling
the enclosed space in and, then, of abandoning the
traditional structure entirely. Such emergent ideas are
typical of innovative thinking, but what are the
implications for computing?
Gerhard Fischer
The power of the unaided, individual mind is highly
overrated - much of our human intelligence and
creativity results from the collective memory of
humankind and of the artefacts and technology
surrounding us. Rather than studying humans in
isolation, we have to develop models of distributed
cognition and new role distributions between humans
and computers. To exploit artefact, group and
institutional memories and to bring design concepts into
unseen and untaught, yet relevant contexts, new
representations are needed to serve the task at hand.
Task-relevant reminding is critical for creative activities.
'Artefacts do often not speak for themselves' - therefore
mechanisms are needed to increase the back-talk of
artefacts. Human knowledge is tacit and it only surfaces
in the context of specific tasks. This implies that
problems are not given, requiring the integration of
problem framing and problem solving. In our research
over the last decade we have tried to create
computational artefacts supporting these challenges. The
domain-orientation of our design environments brings
tasks to the forefront, thereby transcending 'human
computer interaction' by supporting 'human problem-
domain interaction'.
Joy Mountford
Have you ever seen anyone, doing much creative work
directly using a computer? Computers were 'invented'
more as productivity aids, which has made them more
or less useless in the creative domains. When do you
see mathematicians, even, working their thoughts
through directly on a computer screen? The tools of the
creative businesses are typically paper and pencil, white
boards and physical objects. Ideas are initially sketched
out in a rough form. The nice, orderly approach of
word processing makes everything 'look' and be
perceived in a finished form. The issue is how much
can we change this orderly system to encourage the use
of the computer for alternate methods and thought
processes? Is it as simple as change the I/O
environment? What is important to capture during the
process of creation?. What are the relevant parts of the
entire process, and when are these different from
performing or answering a problem? I think the
computing domain needs a balancing of their focus
areas, to those that enable more creative acts. This is
only likely to occur when artists and scientists are more
aggressively encouraged to work much closer together.
Frieder Nake
The Hungarian composer of complex music ("maximal
music"), György Ligeti, says he is making minimal,
close to zero, use of the computer, but maximal use of
his brain. The computer does not, by itself, influence
creativity much. Yet its existence changes our views of
the world, and thus it has an impact on creative work. It
has often been said that with the computer, artists,
designers, or architects may easily play with hundred of
variations. Thus, computers have an impact on the
combinatorial aspects of creativity. Combinatorics,
however, is only the trivial aspect of creativity. Saying
this, should not divert us from the importance of the
groundwork of creativity. The computer is "the machine
to think with", it has been said. More precisely, it is a
semiotic machine, the machine to carry out algorithmic
semioses (sign processes). If we want to understand the
relation of computers to creativity, we can learn from
conceptual art. Creativity happens when an innovative
idea encounters the proper material, and "shapes" it.
When creating, I may directly manipulate my material,
or I may only describe how to manipulate it. A definite
influence of computers on creativity is the separation of
description and manipulation. The computer is the
machine for creativity in post-modernism.
Douglas Riecken
How might the process of innovative design benefit
from the application of computing technologies? We
could employ the computer during a design session to
enumerate an exhaustive set of views representing
plausible solutions to a given problem. Of course, this
would require the computer to be endowed with an
extensive domain specific knowledge-base. A critical
concern regarding this approach is the embodiment of
aesthetics in the knowledge-base. When formulating
solutions, a knowledge-based approach could focus its
search criteria biased by emergent design properties
which satisfy some aesthetic value. To achieve this
behaviour, a knowledge-based system should (in the
minimal case) functionally entail two physical
characteristics. First, the knowledge schema which
serves to represent various compositional levels,
ranging from the design primitives to complex
composite design artefacts, must be extremely flexible;
the schema must be reconfigurable so as to functionally
support plan reformulation. The knowledge embodied
in the system must provide a representation for
aesthetic values and a mapping between these values
and the different types of design artefacts.
Robert Spence
I have recently completed a series of interviews with
visionary engineers in a project which has attempted to
look forward to the design office in the year 2020. A
topic of wide concern, and commented upon in depth, the
early 'creative' stage of design was nevertheless one for
which a pressing need was identified rather than solutions
envisaged. The possibility of the computer itself being
creative was dismissed. Pencil (probably soft) and paper,
together with the essential eraser, frequently combined
with face-to-face discussion with colleagues, will still be
common in 2020. What will have emerged by then,
however, is a range of ways in which the computer will
facilitate the creative process. Emergence, where a new
concept emerges from the combination of two existing
ones, may be facilitated by pattern recognition and/or
neural networks. And CAD software will be so designed
as to allow suspension of judgement so that decisions can
be made at any time. The form of computer-based tools
will reflect the need to support two concurrent processes,
those of problem formulation and problem solution
proceeding in tandem, at any level from component to
system.
References
- Boden, M.A. The Creative Mind. Weidenfeld and
Nicolson, London, 1990.
- Dartnall, T (ed) Artificial Intelligence and Creativity.
Kluwer Academic, Dordrecht, 1994.
- Edmonds, E.A. Culture, Knowledge and Creativity -
Beyond Computable Numbers, Languages of Design, 1,
3, pp 253-261, 1994.
- Fischer, G. Turning Breakdowns into Opportunities for
Creativity. Knowledge-Based Systems, pp 221-232, 1994.
- Gero, G & Maher, M-L (eds), Modelling Creativity and
Knowledge-Based Creative Design, Erlbaum, Hillsdale,
New Jersey, 1993.
- Schön, D.A. The Reflective Practitioner, Maurice
Temple Smith, London, 1983.
