Noise of many

Most systems designed for collective participation make individual action legible. Noise of Many inverts this. Each click changes a single pixel in a shared image permanently, but in almost all cases invisible to the person who made it. The image belongs to everyone who contributed and resembles no one, yet everyone.

The work started from a specific observation: online participation has largely been reorganized around consumption, and platforms track and monetize individual behavior while keeping collective structure opaque. The question was whether a system could do the opposite, making collective structure visible while the individual contribution dissolves into it.

Built as a web application backed by a real-time database, every click is stored and translated immediately onto a shared canvas, marked by a brief visual flash before dissolving into the accumulated image. With no author and no moderator, the canvas records aggregation rather than coordination. What accumulates is itself a form of noise, an order that emerges from statistical superimposition.

Rancière’s concept of the distribution of the sensible shaped the framing: what gets to appear within a shared perceptual field? A single click is statistically negligible. In aggregation, however, these clicks compose something. The work lives in that gap between individual action and collective effect. Mouffe’s writing on democratic paradox adds a second frame: democratic participation does not converge toward consensus. The image Noise of Many accumulates reflects this. It is not legible as a collective decision.

A prototype is accessible at eeeh.fun.

Kaleidoscope

The piece was built around a simple premise: that people might move again as children do, out of curiosity rather than self-consciousness.

Two cameras track the dancer through optical flow. A TouchDesigner system translates movement into abstract projections: lines, flowing gradients, sudden fields of color. The setup forms a closed feedback loop. The dancer moves, the projection responds visibly, the dancer reacts to the response. The projection is not background. It is the material the performer works with in real time. This shifts what the system is asked to do. It does not interpret movement or render it accurately. It produces visual events that are legible enough for the performer to pick up on and unpredictable enough to prompt the next movement.

The loop runs in both directions: unfamiliar movement shifts perception, and shifted perception invites unfamiliar movement. The same dynamic became visible in the audience. The performance opened deliberately to participation, and visitors moved into it. Most of them started cautiously, then noticed the projection responding, and let that response carry them further than they would have moved on their own.