Information Technology Reference
In-Depth Information
6.5
Relational Properties in Ubimus Design:
Methodological Implications
Given the radical shift set forth by the interaction aesthetics and the ubiquitous
music research programs, existing approaches to interaction design are being refor-
mulated bringing new issues to the foreground. Relational properties - such as pli-
ability (Löwgren
2009
) and anchoring (Keller et al.
2010
) - capture the tight
agent-object interaction dynamic that has been at the center of the embedded-
embodied approaches to cognition (Gibson
1979
). In line with the enactive perspec-
tive (Di Paolo et al.
2010
), these design qualities emerge as a result of mutual
adaptations between agents and objects. Because they demand the active engage-
ment of an agent, relational properties cannot be “attached” to objects. They can
only be experienced “in the act.” There are several methodological implications of
adopting relational properties as theoretical constructs.
Firstly, design experiments must take into account both subjective and material
constraints and opportunities. Experiments that adopt closed epistemic fi elds -
where subjects are given a task to complete in laboratory settings and the results
measure their effi ciency in completing the task - do not provide information on the
ecosystem's support for the emergence of relational properties. In this case, the
material resources are chosen by the experimenter and may not fulfi ll the creative
needs of the subject. This is an example of the problem that ubiquitous music
research has labeled as the “auto-referentiality of the theoretical-experimental con-
struct” (Keller et al.
2011b
): the experimental situation does not take into account
the conditions of real-world settings. Relational properties may be absent from a
laboratory task but may be accessible when participants are interacting on site.
Second, the adoption of a tool at the initial stages of the design process may
forgo the emergence of relational properties. Sometimes, the agent-object ecosys-
tem's potentials to support relational properties are not enough to ensure creative
outcomes. These potentials have to be materialized in products and behaviors.
Given that behaviors are constrained by personal factors (including both cognitive
and social resources), understanding the creative profi les of the subjects is an inte-
gral part of the design process. From a ubiquitous music perspective, this procedural
limitation is defi ned as “early domain restriction” (Keller et al.
2011b
).
Third, assessments purely based on products may not provide a complete picture
of the creative factors at play. There is a rich literature of product evaluation meth-
ods in creativity studies (Baer and McKool
2009
; Mumford et al.
2011
). Creative
products give reliable information on creative outcomes. This information can be
enhanced by the analysis of domain-specifi c products. Nevertheless, despite its
untapped potential, this retrospective approach has an epistemological limitation:
products and procedures are usually not equivalent. Whether considering synchro-
nous or asynchronous creative activities, if a relational property emerges as a by-
product, it may not be discernible through the analysis of the fi nished product but it
may be observable while the product is being made or used. Asynchronous cre-
ative activities furnish procedural data that needs to be collected during the creative
