Daily Note: Exploring Customization in Intelligent Systems

These notes are a summary of concepts presented in “Patterns of Sharing Customizable Software.”

Wendy E. Mackay. 1990. Patterns of sharing customizable software. In Proceedings of the 1990 ACM conference on Computer-supported cooperative work (CSCW ’90). Association for Computing Machinery, New York, NY, USA, 209–221. https://doi.org/10.1145/99332.99356

  1. Customization and Emerging Technologies
    • Historical view: Software customization as a solitary activity for expressing individual preferences (Mackay, 1990).
    • Modern evolution: The rise of intelligent systems and their ability to customize software on behalf of users.
    • Question: How does Human-in-the-Loop (HITL) influence this dynamic?
      • HITL definition: Collaborative integration of human input into ML and AI systems’ lifecycle (Source).
  2. Usability Heuristics and Intelligent Systems
    • Visibility of System Status: Keeping users informed with timely feedback.
    • User Control and Freedom: Emergency exits for undoing unintended actions.
    • Recognition Rather than Recall: Reducing cognitive load with visible or easily retrievable options.
    • Flexibility and Efficiency of Use: Shortcuts for expert users while remaining accessible to novices.
    • Help Users Recognize, Diagnose, and Recover from Errors: Clear, constructive error messages.
  3. Reflective Software
    • Definition: Software capable of runtime self-inspection and modification.
    • Potential link to intelligent system adaptability.
  4. Consumer Platform-Specific Customization Models
    • iOS: Customization via app/widget color, tint, size, and labels.
    • Android: Third-party launchers and pinned shortcuts.
    • Peer influence: Sharing customization ideas and inspiration
  5. User Interaction and Evolution Over Time
    • Impact on daily behavior patterns and adaptability.
    • User archetypes
      • How intelligent systems influence user behavior.
      • How resistance to change is managed or mitigated.
  6. Knowledge Sharing and Customization Metrics
    • Evaluating intelligent system customizations:
      • Effectiveness and user relevance.
      • Metrics for success.
    • Mechanisms for knowledge sharing:
      • Interface features to facilitate:
        • Evaluating customizations.
        • Identifying problems/errors.
        • Supporting feedback and translation roles.
        • Building trust
  7. Incentives for Customization
    • Role of incentives in motivating intelligent system customization.
    • Incentives as a catalyst for user engagement and innovation.
  8. Community and Collaboration
    • Interfaces for group discussions among similar users:
      • Sharing effective customizations.
      • Learning new features.
      • Providing feedback loops.
    • Mechanisms to enhance collaboration
      • Problem identification and resolution.
  9. Scope of Customization
    • Data sources
    • Device features