Back to the real world: Tangible interaction for design

After several decades in which design computing has been almost exclusively the domain of software, today, many investigators are building hybrid systems and tools that in one way or another bridge the divide between physical “real-world” artifacts and computational artifacts. On the one hand, the rise and popularity of mass customization, rapid prototyping, and manufacturing raises questions about the kinds of software systems and tools that will make these hardware technologies useful in designing. In contrast, advances in microcontroller and communications technologies have led to a wave of embedding computation in physical artifacts and environments.Botharedescribed inGershenfeld’s (1999,2005)popular books. Tangible interaction is a growing field that draws technology and methods from disciplines as diverse as human–computer interaction (HCI), industrial design, engineering, and psychology. If the idea of ubiquitous computing (Weiser, 1991) is computation integrated seamlessly into the world in different forms for different tasks, tangibility gives physical form and meaning to computational resources and data. The HCI community terms this seamless interaction variously “tangible bits” (Ishii & Ullmer, 1997) or “embodied interaction” (Dourish, 2001). Tangible interaction is simply coupling digital information to physical representations. A tangible user interface (TUI) extends the traditional graphical user interface on screens into the everyday physical world to realize the old goal of “direct manipulation” of data (Shneiderman, 1983). Aish (1979) and Frazer et al. (1980) were pioneers in tangible interaction for design; both developed instrumented physical objects as input devices for computer-aided design. However, the widespread adoption of the mouse in the first commercial personal computers shadowed other forms of interacting with computers such as the pen and TUIs. The development of TUIs based on augmented reality and embedded computing proceeded independently, and it was many years before the HCI community rediscovered these early efforts (Sutphen et al., 2000). Coupling physical features to digital information can be perceptually mediated by human senses (i.e., sight, touch, smell, etc.), leveraging the affordances of things (blocks are stackable, balls are bouncy) and control mechanisms (squeezing toothpaste out of the tube or turning a knob). Before the days of fast CPUs programmers could read the changing patterns of lights on the mainframe console (known colloquially as “das Blinkenlights”) to help debug their code, an early instance of bringing processes within the machine into the physical realm. More recent tangible interaction research makes virtual objects “graspable” (Fitzmaurice et al., 1995) by using physical “bricks” as handles and letting people “take the digital out” into the real world to manipulate it physically. Blocks, for example, are a popular physical form for tangible interaction. We rotate Navigational Blocks (Camarata et al., 2002) to query digital content in a tourist spot; flip and rotate blocks to scroll a map on Z-Agon (Matsumoto et al., 2006); stack up Computational Building Blocks (Anderson et al., 2000) to model three dimensions; or snap together Active Cubes (Watanabe et al., 2004) to interact with the virtual world. Alternatively, consider projection: projection and spatial augmented reality have been employed with multitouch surfaces and computer vision in reacTable (Jordà et al., 2007) for hands-on musical collaboration or in bringing dinosaurs to life in Virtual Showcase (Bimber et al., 2003). Coming full circle is the “rich interaction camera” (Frens, 2006) that applies the form, interaction, and functionality of a conventional 35-mm film camera to improve the usability of its digital counterpart. Since its first issue, AI EDAM has published the best work at the frontiers of engineering design and computing; today, tangible interaction is one of those frontiers. At first glance tangible interaction might seem a mere conceit: interfaces are inherently superficial. However, new input and output Reprint requests to: Ellen Yi-Luen Do, ACME Lab, College of Architecture and College of Computing, Georgia Institute of Technology, Atlanta, GA 30332-0155, USA. E-mail: [email protected] Artificial Intelligence for Engineering Design, Analysis and Manufacturing (2009), 23, 221–223. Printed in the USA. Copyright # 2009 Cambridge University Press 0890-0604/09 $25.00 doi:10.1017/S0890060409000195