Channel: Stanford Online

Category: Education

Tags: cs547computer sciencestanfordstanford onlinedesign spacesstanford seminar

Description: Tamara Munzner is a Professor of Computer Science at the University of British Columbia. This talk was given March 4, 2022. Design spaces impose a systematic structure on the set of possibilities, intended to capture the key variables at play in the context of a particular design problem. These categorizations, often delineated by cross-cutting axes or dimensions, support the combinatorial enumeration of and reasoning about design choices. I will present several design spaces that focus on visualization concerns. All of them have a common starting point, namely the qualitative approach of thematic analysis, but they diverge with respect to what type of source material was open-coded. In a literature study, we used the source material of analysis reports extracted from previously written papers on application-driven visualization work, called design studies in the visualization literature, to create a framework of high-level abstract analysis goals. In a technical observation study, we used the source material of software artifacts to create descriptive taxonomies of the data wrangling actions and processes of journalists, followed by reflective synthesis to create a design space for multi-table data wrangling. From a systematic survey of the figures in published papers from the genomic epidemiology literature, we derived a typology of how visualizations were created from a series of basic chart types, and how those charts were combined and enhanced. I will discuss the assessment of design spaces according to many criteria, including descriptive power, generative power, resonance, and parsimony. View the full playlist of Stanford Seminars here: youtube.com/playlist?list=PLoROMvodv4rMWw6rRoeSpkiseTHzWj6vu Learn more about our Human-Computer Interaction Graduate Program: online.stanford.edu/programs/human-computer-interaction-graduate-program 0:00 Introduction 0:54 Design spaces: Continuing theme 1:53 Design spaces: What are they! 6:41 Design spaces: How to create! 10:51 Design space with three axes 11:51 Combinations: Characterize narrative, perceptual 12:45 Viable combinations 14:59 Assessment & adoption descriptive power 19:25 Domain prevalence design space 20:44 By the numbers 22:17 Design space axis: Chart combinations of heterogeneous data 23:00 Design space axis: Enhancement choices, atop base chart types 27:20 Process overview 30:39 Key finding journalists use many, many tables 32:02 Multi-table data wrangling design space 32:23 Assessment: Cross-check 33:20 Task abstraction: Gap 35:36 Mapping terms 40:19 Summary: Multiple design spaces 44:47 Design spaces: How to assess? Larger context theory types