<quick_start> When facing a problem, ask:
Instead of: "What agent/assistant can do this for me?" Ask: "What new sense would let me perceive this problem differently?"
The goal is not automation. The goal is augmentation. </quick_start>
<essential_distinction>
| Copilot (Anti-pattern) | HUD (Target) |
|---|---|
| You talk to it | You see through it |
| Demands attention | Operates in periphery |
| Delegates your judgment | Extends your perception |
| Context-switching tax | Flow-state preserving |
| "Do this for me" | "Now I notice more" |
| </essential_distinction> |
<reframing_process> To reframe any problem using HUD-first thinking:
- •
Identify the copilot instinct
- •What task are you tempted to delegate?
- •What conversation would you have with an assistant?
- •
Extract the information need
- •What does the assistant need to know to help?
- •What would you need to perceive to not need the assistant?
- •
Design the sense extension
- •What visual/auditory/haptic signal would make this obvious?
- •How could this information be ambient rather than on-demand?
- •
Validate with the spellcheck test
- •Spellcheck doesn't ask "would you like help spelling?"
- •It just shows red squiggles. You notice. You decide.
- •Does your solution pass this test? </reframing_process>
<hud_approach>
- •Inline complexity warnings (like spell-check for cognitive load)
- •Test coverage heatmap in the gutter
- •Type inference annotations that appear on hover
- •Mutation testing results as background highlights → You see code quality. No conversation needed. </hud_approach> </example>
<hud_approach>
- •Urgency highlighting (color gradient based on signals)
- •Relationship context badges (how often you interact)
- •Sentiment indicators (tone of message)
- •Thread age/velocity visualization → You perceive inbox state at a glance. You decide what matters. </hud_approach> </example>
<hud_approach>
- •Live variable values overlaid during execution
- •Control flow visualization (which branches taken)
- •State diff between invocations
- •Anomaly highlighting (this value is unusual) → You see program behavior. The bug becomes obvious. </hud_approach> </example>
<hud_approach>
- •Readability score in margin (updates as you type)
- •Sentence complexity highlighting
- •Passive voice indicators
- •Repetition detection → You sense where prose is weak. You fix it your way. </hud_approach> </example>
<design_principles> From Calm Technology (Weiser, Case):
- •Require minimal attention — Lives in peripheral awareness
- •Extend senses, don't replace judgment — New information channels, same human decision-maker
- •Communicate without speaking — Color, position, sound, vibration—not dialog boxes
- •Stay invisible until needed — Information surfaces when relevant, recedes when not
- •Amplify Human+Machine — Optimize the interface between them, not either alone </design_principles>
<when_copilot_is_fine> Delegation works for:
- •Routine, predictable tasks (autopilot for straight-level flight)
- •Tasks you genuinely don't want to understand
- •One-time operations with clear success criteria
But for expert work, creative work, complex judgment—you want instruments, not a chatbot to argue with. </when_copilot_is_fine>
<challenge> For your current problem:- •What would a "red squiggly" look like for this domain?
- •What sense would you need to perceive the solution space directly?
- •How could the information be ambient and continuous rather than requested and discrete?
The best AI interface is often invisible. You just become aware of more. </challenge>
<success_criteria> HUD-first reframing is successful when:
- •The proposed solution doesn't require conversation or explicit requests
- •Information flows continuously rather than on-demand
- •The human remains in control of judgment and decision
- •Flow state is preserved (no context-switching to interact with AI)
- •The user would describe it as "now I just notice things I didn't before" </success_criteria>