Supporting the design and fabrication of physical visualizations

Abstract

Figure 1. Physical visualizations created with our fabrication tool, MakerVis: a) a layered scatterplot modeled after Rosling ; b) a prism map showing relative happiness in US states computed from Twitter sentiments; c), d), e) layered bar charts and line charts crafted by end users.

Opportunities and Challenges for Data Physicalization

“…Early on, Vande Moere and colleagues pointed out the potential of data sculptures as a medium and the need to study them more deeply [43,75,79]. Later, several papers were published in Visualization and HCI venues that investigate empirical [34], design [67,69], and conceptual [32,33,75,79] questions regarding physicalizations, as well as specific application domains [68]. However, the physicalizations investigated are static and none of this work involves novel input and output technologies.…”

“…Research has already demonstrated the process for composing and creating one-off static physicalizations using fabrication technologies [68,69]. Dynamic physicalizations, on the other hand, require computer-driven control of physical geometry or of material properties.…”

“…Researchers have begun to address this challenge by applying 2D and 3D visual encodings to physicalizations [69]. However, classical visual encodings largely ignore the multisensory nature of physical exploration that we previously highlighted as a benefit.…”

“…Digital fabrication technologies greatly facilitated the process of prototyping and creating static physicalizations [68,69] but even the most elaborate digital fabrication machines available today have not been built with data physicalization in mind. Multi-material 3D printers can interleave different source materials to gain some form of control over properties such as color, opacity, or deformability, but the range of materials available is still very limited [25,47].…”

“…Finally, prototyping and building physicalizations involve complex workflows, ranging from laborious and low-fidelity manual crafting to custom programming (less laborious but requires expertise) to fully automatic (lower entry requirements but little control over the design) [69]. Thus, one challenge is to characterize possible fabrication workflows and classify them as to their benefits and drawbacks, including how fast physicalizations can be created or replicated, how accurate they are, how expensive the required material is, their environmental impact, and which skills are required to create them.…”