Professor Jeffrey Heer, who leads the Allen School’s Interactive Data Lab, and former student Dominik Moritz (Ph.D., ‘19), who was co-advised by Allen School adjunct professor and information science professor Bill Howe, were each honored for their impact on interactive visualization research at IEEE VIS 2020 this week, the flagship conference in the field of visualization and visual analytics. Heer received the IEEE InfoVis 10-Year Test of Time Award for his 2010 paper, “Narrative Visualization: Telling Stories with Data,” while Moritz received the IEEE Visual Graphics and Technical Community (VGTC) Doctoral Dissertation Award for his thesis “Interactive Systems for Scalable Visualization and Analysis.”
In the winning Test of Time paper, Heer and co-author Edward Segel explored how visual data enhances journalistic storytelling and studied design strategies for narrative visualization. The paper helped to frame and advance research into the use of visualization for journalistic reporting and storytelling. Since then, it has been widely cited and influential in the fields of both visualization and data-driven journalism.
Fascinated by the growing use of visualizations in online journalism, Heer and Segel built a catalog of examples to identify distinct genres of narrative visualization. The two characterized the design differences and messaging and found that many samples could be more dynamic with the help of more sophisticated online tools — including those that allow interactive exploration by the reader.
When the paper was originally published, Heer was a professor of computer science at Stanford University and Segel was a master’s student. Together, they created a comprehensive framework of design strategies for narrative visualization.
“We wanted to better understand the innovative work of data journalists and designers whose insights we hoped to give further reach with our paper,” Heer said. “From the framework of our research, we found promising yet under-utilized approaches to integrating visualization with other media, and the potential for improved user interfaces for crafting data stories.”
Heer had already started to develop a series of robust tools for producing interactive visualizations on the web. As a graduate student, he helped to create Prefuse, one of the first software frameworks for information visualization, and Flare, a version of Prefuse built for Adobe Flash that was partly informed by his work in animated transitions. This latest research with Segel focused on a central concern in the design of narrative visualizations: the balance between author-driven elements that provide narrative structure and messaging, and reader-driven elements that enable interactive exploration and social sharing. This work helped to identify successful design practices that guided the development of new narrative visualization tools.
Since joining the Allen School faculty in 2013, Heer has worked on a suite of complementary tools for data analysis and visualization design built on Vega, a declarative language for producing interactive visualizations. These tools include Lyra, an interactive environment for generating customized visualizations, and Voyager, a recommendation-powered visualization browser. In 2017 he was recognized with the IEEE Visualization Technical Achievement Award and the ACM Grace Murray Hopper Award for his significant technical contributions early in his career.
Vega led to Vega-Lite, a project that earned Heer and Moritz — now a professor in the Human-Computer Interaction Institute at Carnegie Mellon — a Best Paper Award at InfoVis 2016 along with their collaborators. Vega-Lite is a high-level grammar for rapid and concise specification of interactive data visualizations. The goal was to enable non-programmers to create sophisticated visualizations that can be generated automatically. That project and others formed the basis of Moritz’s 2019 dissertation, which made a number of contributions spanning formal languages, automatic reasoning for visualization design, and novel approaches for scaling interactive visualization to massive datasets for which he was honored at this year’s VIS conference.
One of those contributions was Draco, an open-source, constraint-based system that formalized guidelines for visualization design and their application in visualization tools. The system, which earned Moritz and his colleagues a Best Paper Award at InfoVis 2018, offers a one-stop shop for researchers and practitioners to apply and test a set of accepted design principles and preferences and to make adjustments to their visualizations based on the results. To expand the application of user-friendly visualization tools to larger datasets, Moritz introduced Pangloss, which enables analysts to interactively explore approximate results pending completion of long-running queries. Pangloss generates visualizations based on samples while queries are ongoing, with the ability to detect and correct errors later. Moritz followed that up with Falcon, a web-based system that supports real-time exploration of billion-record datasets by enabling low-latency interactions across linked visualizations.
Moritz’s interactive systems for visualization and analysis have seen widespread adoption in the Python and JavaScript communities.
Congratulations Jeff and Dominik!