Ten Million Users and Ten Years Later: Python Tutor's Design Guidelines for Building Scalable and Sustainable Research Software in Academia

Document Title

Ten Years Later, the Story of Ten Million Users: Design Guidelines for Building Scalable and Sustainable Research Software in Academia with Python Tutor

Document Information

• Subject Area: Human-Computer Interaction (HCI), Software Engineering, Programming Education Tools
• Keywords: Research Software, Sustainability, Python Tutor, Code Visualization, User Experience, Software Architecture, Academic Software, Programming Teaching Tools, Program Visualization, Software Design Principles

Research Background and Problem

• Problems or Challenges Identified by the Authors:
  • Many academic research software projects are developed as proof-of-concept tools. Despite their potential for widespread application, they are often abandoned due to lack of maintenance.
  • Resource constraints in the current academic environment (e.g., unstable developer positions, lack of long-term funding) make building sustainable research software particularly challenging.
• Why This Problem Is Important:
  • Sustainable research software can achieve broader user adoption and help academic research generate long-term impact.
  • Software can provide platforms and resources for other research communities, supporting subsequent studies.
• Research Motivation and Related Work:
  • The authors aim to distill a set of general design principles for developing scalable and sustainable software in academia, based on their decade-long experience in developing and maintaining Python Tutor.
  • Related work has discussed program visualization tools and large-scale academic online experiments (e.g., LabintheWild), but the authors' systematic summary of contributions in technical design and sustainability is novel.

Solution

• Proposed Solution:
  • The authors introduce Python Tutor (a widely-used code execution visualization tool) and derive ten specific design guidelines across three core design domains:
    • Principles for User Experience Design: Ensure the tool is easy to use, requires no installation, and does not mandate user login.
    • Principles for Software Architecture Design: For example, maintain statelessness, use older technologies, and minimize external dependencies.
    • Principles for Software Development Workflow: For instance, assign a single developer, focus on specific needs, and ignore most user suggestions to prevent feature bloat.
• Innovative Aspects of the Solution:
  • Emphasis on the unique constraints of academic software (limited resources and developers) and distinctions from industry best practices.
  • Strategies such as avoiding "login walls" and promoting a "no storage content" policy to minimize user attrition and reduce long-term maintenance burdens.
• Implementation Steps and Key Technologies:
  1. Build the infrastructure using lightweight technologies (e.g., older versions of JavaScript, CGI).
  2. Implement a stateless architecture, storing all states in the user's browser.
  3. Provide URL-sharing links to allow users to embed and distribute visualization content.
  4. Use Docker containerization to achieve reproducible environments.

Research Outcomes

• Specific Achievements:
  • Python Tutor has been used by over 10 million users across 180 countries, playing a significant role in education and online communities.
  • To date, the tool has supported 55 academic publications, contributing to the advancement of programming education research.
• Advantages Compared to Existing Solutions:
  • Unlike many academic software projects that are abandoned after brief usage, Python Tutor has successfully supported users for over a decade.
  • The authors avoided content moderation pressures (e.g., by enabling visualization content to be embedded and shared on other websites), demonstrating the success of the "share rather than store" strategy.
• Experimental or Evaluation Results:
  • The authors estimated Python Tutor's academic impact through Google Scholar, finding its code reused by multiple research groups and supporting various user behavior studies.
  • Completed user surveys indicate widespread recognition of the platform's usability.
• Limitations and Future Directions:
  • Limitations: The "bus factor" of a single-developer model (if the developer stops maintaining, the project may stagnate); potential performance limitations due to the choice of older technologies.
  • Future Directions: Explore ways to promote the design principles summarized by the authors in broader academic contexts, and advocate for institutional support to drive long-term research software development.

Additional Summary

This article serves as both a comprehensive review of the Python Tutor tool and a practical guide for academia. The design principles distilled from the authors' decade-long experience in development, maintenance, and promotion can be referenced by other research teams to achieve broader impact under limited academic resources.