Information Technology Reference
In-Depth Information
the performer. This is a particular application of system G and H in Fig.
6.1
.De-
signers can place the greatest emphasis on the design and behaviour of
f
, exploring
the musical behaviour of diverse computational systems, possible with a flexible
approach to the form of
P
and
Q
. Through such mutual influence, the performer
and Live Algorithm can be seen as a coupled dynamical system, where both partic-
ipants are capable of acting independently. Coupling does not prescribe a specific
behaviour, and may involve aspects of mirroring and shadowing (in the latter case
the coupling would be tighter), but tends to refer to a situation in which the system
can clearly be left to lead (by acting more independently of the performer), possibly
to the detriment of the sense of participation (in which case we can think of the algo-
rithm as “stubborn” or unresponsive). However, a sense of participation depends on
the attitude of the performer. A performer may deride shadowing and mirroring as a
failure to truly participate, that is, to bring something original to the collective per-
formance. A successful coupling-based system would demonstrate autonomy and
creativity, and in doing so achieve participation.
Coupling is a practical behaviour because it is essentially trivial to achieve; it
evades strict requirements about the kind of interactive behaviour the system ex-
hibits, as long as the performer is demonstrably exerting some kind of influence
over the system. This offers creatively fertile ground for what McLean and Wig-
gins (
2010
) refer to as the
bricoleur programmer
, building a Live Algorithm by
musical experimentation and tinkering. It also relates to an aesthetic approach to
computer music performance that favours autonomy, leadership and the potential
for surprising variation (novelty and thus creativity) over participation. It allows for
the introduction of various generative art behaviours into a performance context.
6.4.1.4 Negotiation
Negotiation is defined as a more sophisticated behaviour that is related to coupling
but is based on aspects of human cognition. Only system H in Fig.
6.1
allows for this
behaviour. A system that negotiates constructs an expectation of the collective mu-
sical output and attempts to achieve this global target by modifying its output. Since
the collective musical output depends on the performer as well, negotiation, as the
name suggests, may involve attempts to manipulate the behaviour of the performer,
or equally, to adjust one's expectations in light of the direction of the music. As with
coupling, with negotiation the Live Algorithm is understood as interacting directly
with a piece of music, rather than with other individuals. More sophisticated Live
Algorithms could perform acoustic source separation and use a “theory of mind” to
infer individual behaviour from the environment.
Negotiation can be seen as a framework for the design of a Live Algorithm (see
Young and Bown
2010
), involving the interaction between an expectation and a
behaviour (which contributes to the musical environment), either of which can be
modified to create a better fit between them. This is harder to achieve than the other
behaviours, so is less pragmatic. The challenge is to obtain a system that achieves