Pablo Marques

Point Cloud — by James Leng

Point Cloud is an attempt to reimagine our daily interaction with weather data. Weather has always had a unique place in our lives, because it has a multiplicity that encompasses both the concrete and the indeterminate. It is the intangible context within which we build our lives and our cities, but it is also the physical element against which we create protective shelter. Most of the time it is an invisible network that we can see but are not aware of; yet it can manifest in a spectacle or disaster, come forward and activate our senses, make us forget our rationality in delight or fear. With modern scientific and technological developments, we can now deploy sophisticated monitoring devices to document and observe weather. Yet despite these advances, our analysis and understanding of meteorology is still largely approximate, and in many cases, inaccurate. Weather continues surprise us and elude our best attempts to predict, control, and harness the various elements.

In contrast, however, the nuances of weather’s continuously shifting states are largely oversimplified as the information is transmitted into our daily experience. Our various home and mobile devices most likely distill a forecast into static representations, such as numeric values or simple infographics of sun, clouds, or rain. There is a deep discrepancy between the flatness of the visualizations we are accustomed to, and the rich mixture of tactility and perceptibility of our immediate physical experience. As a critical response to these issues, Point Cloud emerges as a sculptural form defined by a thin wire mesh, driven asynchronously by 8 individual servos controlled via Arduino. As whiteness of the hanging structure begins to disappear into the background, the viewer is treated to a constantly morphing swarm of black points dancing through midair.

The project’s ambitions are two-fold - first and foremost it seeks to meditate on the transmutation of digital data back into analog movement. In the current prototype, the speed, smoothness, and direction of rotation are modulated to interpret a live feed of weather data. Instead of displaying static values of temperature, humidity, or precipitation, Point Cloud performs the data, dynamically shifting between stability and turbulence, expansion and contraction. It re-introduces weather conditions as a permanently variable state, and creates a visceral experience in our interactions with weather.

The second aspiration of the project is an attempt to conceptually mime the structural complexity of meteorological systems - wherein predictable elements converge into unpredictable or unexpected outcomes. The various components of Point Cloud are functionally autonomous and clearly defined: 4 cables descend into a central control core, from which a lightweight steel space frame cantilevers and supports the 8 servos. Each servo powers a cam mechanism that activates 3 to 4 pistons that push and pull on various parts of the wire mesh - composed of over 300 feet of wire thread, and 966 intersection joints. Despite the fact that the only type of mechanical actuation is linear, the resulting motion is like that of a third degree digital surface; the effects each push and pull ripple out along the elastic tension of the wire threads, and in combination with the syncopated rhythm of the servos, create movement that is complex, unexpected, and hopefully wondrous.

Video of earlier Prototypes: vimeo.com/38959713
Photos & Process: flickr.com/photos/ettubrute/sets/72157629909864772/