Reciportrait: a Data Humanism Approach for Collaborative Sensemaking of Personal Data

Research Background and Issues

• Identified Problems and Challenges: Current personal data visualization tools often focus on objective quantification and simplified data presentation, neglecting the subjectivity and narrative aspects of data. While "data humanism" advocates for a slow-paced, in-depth approach to analyzing personal data, collaborative sensemaking introduces additional structural and coordination requirements, creating a conflict with the goals of data humanism.

• Significance: Personal data is increasingly used for behavioral reflection and habit improvement. However, individual analysis alone often fails to uncover hidden behavioral patterns, whereas collaboration can provide diverse perspectives and more comprehensive self-awareness.

• Research Motivation and Related Work: This study is based on the theoretical background of data humanism and collaborative sensemaking, aiming to design tools that balance the needs of both approaches. Building on the limited existing research on collaborative personal data visualization (e.g., data physicalization and participatory visualization), it seeks to address the issue of promoting collaborative sensemaking through personalized yet structured visualization tools.

Solution

• Method or Solution: The authors propose four design principles to balance the demands of data humanism and collaborative sensemaking in a collaborative environment and developed a collaborative personal data visualization tool called "Reciportrait."

• Innovations:

  • Balancing subjectivity and structure in a slow-paced process, coordinating individual analysis and collaborative sensemaking.
  • Implementing a "visualization grid" and transparent overlay mechanism to create flexible individual and shared workspaces.
  • Combining data reprocessing with "manual drawing," enabling users to gain deeper reflection and insights.

• Implementation Steps and Key Technologies:

  1. Propose four design principles: support personalized visual encoding (DP1), guide visualization creation (DP2), enable slow-paced interaction (DP3), and provide collaborative and independent workspaces (DP4).
  2. Develop the Reciportrait tool, including visualization grids, example cards, and data reflection canvases.
  3. Evaluate the effectiveness of these principles and tools through experimental observation studies.

Research Outcomes

• Specific Results:

  • All participants (8 pairs of students, two per group) successfully created three sets of collaborative visualizations, showcasing diverse content and reflecting multi-perspective data analysis.
  • Users co-created visualization methods through the tool, emphasizing subjective perspectives in data comparison while uncovering hidden behavioral patterns and life experience factors.
  • The study identified four types of personal insights gained by users: data patterns, behavioral patterns, experiential context, and self-awareness.

• Advantages Over Existing Solutions:

  • Successfully integrates the slow-paced approach of "data humanism" with the efficiency demands of collaborative sensemaking.
  • Creatively employs transparent grids and shared canvases to facilitate data narratives and dynamic dialogues.

• Experimental and Evaluation Results:

  1. User Feedback: Participants in the experiment reported enhanced understanding of the data through slow and in-depth drawing and discussion.
  2. Insight Moments: Users discovered 30 deep personal insights using the tool, including interpretations of data trends and behavioral patterns, recontextualization of situational factors, and deeper self-awareness.
  3. Specific Examples: For instance, one participant identified excessive usage behavior in certain apps through hand-drawn charts and reflected on this after comparing with their partner.

• Limitations and Future Directions:

  • Limitations:
    • The tool's degree of personalization is limited, and existing templates may not meet all users' practical needs.
    • The participant sample is skewed toward design and engineering fields, potentially limiting the generalizability of the findings.
    • Initial mobile data visualization examples may lead to design fixation, influencing output designs.
  • Future Directions:
    • Add more personalized templates and data examples to enhance the tool's applicability and flexibility.
    • Expand the sample pool to include a broader user base, such as ordinary users with low data literacy.
    • Explore collaborative visualization applications in more complex data domains, such as healthcare or financial data.

In summary, Reciportrait successfully demonstrates how the integration of data humanism and collaborative sensemaking can lead to the design of personal data visualization tools with academic innovation and practical significance. This not only provides valuable insights for future tool design but also inspires deeper discussions on collaboration and data subjectivity.