Videos of InteractionsPlease click images to play videos |
|
![]() |
Interactions is initialized as a blank slate. Audience participants input any web address and Interactions will download images from the website which comprise the visual material for both artist agents. Additionally, keywords from the piece's library are sought on each website, and if a match is found that keyword and corresponding soundclip is used in each agent's aural collage. |
![]() |
Once content has been downloaded by the two agents, their differences in presentation become more apparent. Audience participants vote their preference for one agent or the other by typing on the keyboard or another external device provided with the piece. If an one agent gets a preference vote it preserves its state while the second agent will modify its values to imitate the first. If the losing agent's style grows too close to the valued one, the valued artist will radically modify its approach. Thus the two agents are perpetually chasing one another as directed by the audience's preferences. |
![]() |
This agent model leads to an ever evolving aural and visual collage. Notice as the agent on the right suddenly alters its means of presentation, and by the end of the clip the left agent is mimicing its style. |
![]() |
Here is another moment from the installation. Again, note that the artist agents diverge from one another and eventually evolve toward a similar state. |
| back to Interactions main page | |