DIS • Designing Interactive Systems
Uncovering Ambiguity in Crowdsourced Descriptions of Motions and Materials
Although a picture, GIF, or tutorial video is "worth a thousand words", every viewer has a unique perspective of which words they ‘see.’ Packed with rich meaning, these words, or descriptions, are often used to describe the ambiguity of working with materials, artifacts, and processes in creative practices. Our work reframes the traditional crowdsourced description tasks to leverage crowdworker diversity in generating a corpus of tacit descriptions. These descriptions are synthesized into a typology and describe how users communicate ambiguity and tacit knowledge embedded in unfamiliar material properties and familiar handed motions. We propose three design concepts that demonstrate how tacit descriptions could enhance the interpretation of tutorial artifacts - Tacit Layer, or breadth-first interpretation, Tacit Space, or depth-first interpretation, and Tacit Localizer, or context-based interpretation. These concepts are then translated into physical card-sort activities and used in a workshop to gauge how participants consider tacit information in their respective practices.