A framework for spatial interaction in locative media

From Cyborg Anthro Wiki

Authors

Atau Tanaka, composer, performer and researcher in music technology, now Professor of Media Computing at Goldsmiths, University of London; and Petra Gemeinboeck, media artist and researcher working on interactive installation and machine embodiment. Both were at Sony Computer Science Laboratory Paris and associated institutions at the time of writing.

Citation

Tanaka, Atau, and Petra Gemeinboeck. "A Framework for Spatial Interaction in Locative Media." Proceedings of the 2006 International Conference on New Interfaces for Musical Expression (NIME06), Paris, 2006, 26–30.[1]

Description

The paper addresses a problem the locative media movement had largely left implicit: what, precisely, is the interaction model when the input is a body moving through space?

Most locative work of the period used location as a trigger — cross a boundary, receive content. Tanaka and Gemeinboeck argue that this treats space as a set of switches and wastes most of what position actually offers. Their framework proposes instead that spatial data be treated as continuous control, in the sense that a musical instrument treats gesture: position, velocity, direction, proximity and history are all continuously varying parameters that can be mapped to continuously varying output.

The argument comes from the authors' background. In computer music, the question of how to map a performer's gesture to sound is a well-developed field with a substantial literature on the difference between one-to-one, one-to-many and many-to-one mappings, and on why simple mappings produce uninteresting instruments. Tanaka and Gemeinboeck's contribution is to import that literature into locative media, where the equivalent question had not been asked.

Key Ideas

  • Location as continuous parameter, not trigger. The central move. A trigger yields a binary event; a continuous mapping yields an instrument that can be played.
  • Mapping strategies. Applying the NIME tradition's analysis of gestural mapping to spatial data, including the finding that indirect and many-to-many mappings support expressive engagement better than direct ones.
  • Collective and shared space. The framework accommodates several participants whose relative positions are themselves parameters — a group as an ensemble rather than as separate users.
  • The instrument model of place. The implicit proposal that an annotated environment can be something a person plays rather than something that plays at them, which reverses the passivity of most location-triggered content.

Relevance to Cyborg Anthropology

The paper is a rare piece of locative media writing that is about interaction design rather than about theory or art criticism, and its central distinction — trigger versus continuous parameter — is a useful test to apply to any location-aware system. Nearly all commercial location services are trigger-based, which is why they feel like being notified rather than like doing something.

It also connects to sonic metadata: both propose that a continuously varying quantity such as proximity is better rendered as a continuously varying sound than as a discrete alert.

References

  1. ↑ A Framework for Spatial Interaction in Locative Media, NIME 2006 proceedings.
Retrieved from "https://cyborganthropology.com/index.php?title=A_framework_for_spatial_interaction_in_locative_media&oldid=7247"