Multimodal outputs

Agenda

• Overview
• Features and impact
• Empathy map
• Proposed implementation
• Conclusion

Overview

• Adding multimodal output paradigms to developer utilities
• Initial focus on sound-based output for CLI tools in the Scientific Python ecosystem, starting with spin
• Aim: improve developer productivity and accessibility
• Extra: potential community-driven customisation

Features and impact(s)

• Sound notifications for command completions (success/failure/et cetera)
• Customisable sound themes (e.g., 80s retro sounds)
• Various audio formats (.ogg, .wav, .midi, and more)
• Improved accessibility for users with special needs
  • Users with ADHD by providing auditory cues
  • Partially or fully blind users (in conjunction with text-to-speech tools)
• Enhanced developer experience (DX)
• Community engagement through shared configurations

Empathy map

Proposed implementation

• Technology stack: Python, CLI libraries, maybe SDL2

• Potential approaches:

  1. Platform-specific terminal commands

  2. CPython extension linking to SDL library

  3. CLI wrapper over existing PyPI sound projects

• Storage: user-configurable

• Customization via pyproject.toml or environment variables

• Estimated development time: 80-120 hours

• Minimal dependencies, no risks, and no cloud resources needed

Conclusion

• Improving accessibility in technology

• Exploring a minimal Audio User Interface (AUI) for CLIs

• Integrations with popular tools like spin

• Paves the way for more output formats
  (OS level notifications, maybe?)

• This is driven by the sheer fun that comes with writing software