TL;DR
This paper surveys the burgeoning field of storytelling in visualization, offering a novel classification scheme that highlights both mature and emerging research areas. By organizing existing literature based on "who," "how," and "why" storytelling is used, the authors provide a valuable roadmap for researchers and practitioners looking to understand and advance the integration of narrative into data visualization.
What Problem Does This Paper Address?
In an era of data saturation, effectively communicating insights is paramount. Storytelling has long been a powerful tool for conveying information, and its application in visualization is rapidly gaining traction. However, the field is fragmented, with a growing body of research lacking a cohesive structure. This paper addresses the need for a comprehensive survey and a systematic classification of the literature on storytelling in visualization. The authors aim to provide a clear overview of common elements, identify challenges, and offer a starting point for future research by categorizing existing work.
Methods
The authors conducted a thorough literature search across major digital libraries (IEEE, ACM) and visualization publication databases, using keywords related to storytelling, narrative visualization, memorability, transitions, user engagement, and interpretation.
Their primary methodological contribution is a novel classification scheme for storytelling in visualization. This scheme is built upon fundamental questions:
- Who: Who are the main subjects involved? (Authoring-tools, User-engagement)
- How: How are stories told? (Narratives, Transitions)
- Why: Why is storytelling used? (Memorability, Interpretation)
These dimensions are further broken down into sub-categories, such as linear vs. user-directed paths for narratives, and static vs. animated transitions. The paper then categorizes existing research papers based on their primary focus within this framework, presenting the results in a comprehensive table. They also acknowledge that some papers span multiple categories.
Datasets
Not discussed in the paper. The paper is a survey and classification of existing literature, not an empirical study with its own datasets.
Key Insights and Ideas
- The "Who, How, Why" Framework: This classification provides a structured way to understand the diverse landscape of storytelling in visualization, moving beyond ad-hoc categorizations.
- Imbalance in Research: The survey highlights a significant imbalance, with much more research focused on information visualization than on scientific or geospatial visualization.
- Exploration vs. Presentation: A recurring theme is that current visualization techniques tend to prioritize data exploration over presentation, suggesting a need for more work on narrative storytelling for communication.
- Unsolved Challenges: The paper explicitly identifies several critical but under-researched areas, including objective measures of user engagement and data preparation/enhancement for storytelling.
- The Importance of Transitions: Transitions are deemed so crucial that they warrant their own dedicated section, underscoring their role in creating coherent and engaging narratives.
Key Results
The paper's main "result" is the comprehensive survey and classification of the storytelling visualization literature. Key findings include:
- A detailed breakdown of research across categories like Authoring-Tools, User-Engagement, Narratives, Transitions, Memorability, and Interpretation.
- Identification of mature sub-fields (e.g., linear narrative visualization) and nascent areas (e.g., scientific and geospatial storytelling, user engagement metrics).
- A clear mapping of numerous research papers to their respective categories, providing a valuable bibliography for the field.
Limitations and Future Work
Limitations: * The authors acknowledge that the classification is based on their determination of the "main focus" of each paper, and some papers could fit into multiple categories. * The paper focuses primarily on visualization within specific domains (scientific, information, geospatial) and excludes work from fields like Virtual Reality, Augmented Reality, Education, and Gaming, which might offer relevant insights.
Future Work: The paper explicitly outlines several "unsolved problems" and future research directions:
- Measuring User Engagement: Developing objective measures and a mature classification of user engagement activities in visualization.
- Data Preparation and Enhancement: Addressing the challenges of preparing and enhancing data specifically for storytelling purposes.
- Narrative Visualization for Scientific and Geospatial Data: Bridging the research gap in these areas.
- Transitions for Scientific Visualization: Further exploring the benefits of static vs. dynamic transitions.
- Memorability: Identifying key elements that make visualizations memorable.
- Animated Transitions for Geospatial Visualization: A surprising gap given the importance of geospatial data.
- Interpretation for Scientific and Geospatial Information: Further research is needed in this largely unexplored area.
- Standardization: Investigating the possibility of standardized data formats or software systems for storytelling visualization.
Summary updated