Blink: AI-Native Spatial Notes

From static documents to intelligent, spatially aware notes.

---

Vision

Blink reimagines note-taking as AI-native knowledge management. Notes aren’t static text but autonomous agents that can think, update, and contribute on their own.

Core Principles

• Spatial context: where a note lives matters
• Asynchronous work: AI processes notes even when you’re not active
• Conversational notes: interact with each note as an ongoing dialogue
• Emergent organization: structure comes from AI understanding, not manual sorting
• Multimodal inputs: notes handle text, images, voice, and more

---

Current Features

Multi-Window Spatial Notes

• Detach notes into frameless, floating windows
• Spatial arrangement saved and restored
• Real-time sync between main app and detached windows
• Global shortcuts, focus mode, command palette

Customization

• Markdown-first editor with preview
• Font, spacing, and theme controls
• Window transparency, always-on-top, typewriter mode
• Syntax highlighting for code

Core Management

• Instant search with fuzzy matching
• Auto-save with backups
• Smart title extraction
• Cross-platform support

---

Tech Stack

• Frontend: React, TypeScript, TailwindCSS, Zustand
• Backend: Rust (Tauri v2)
• Storage: JSON file-based, migratable to databases
• Build: Vite for dev and cross-platform desktop builds

---

Roadmap

Phase 1 (Done)

• Multi-window spatial foundation
• Command palette, focus mode, customization

Phase 2 (Next)

• AI service integration (Claude, OpenAI)
• Per-note AI context and multimodal inputs
• Background processing

Phase 3

• Conversational notes with history
• Cross-note awareness
• Spatial context in reasoning

Phase 4

• Self-updating notes
• AI-generated insights and briefings
• Pattern recognition across notes

Phase 5

• Note-to-note AI conversations
• Ecosystem-level intelligence
• Collaborative knowledge evolution

---

Architecture

• Spatial-first: window positions as metadata, clustering for related notes
• AI-native: each note has its own AI context and memory
• Local-first: encrypted storage, optional sync
• Rust backend for performance and security

---

Contributing

Areas for contribution:

• AI integration
• Spatial clustering and intelligence
• Multimodal processing (voice, images, docs)
• Rust optimizations

Principles:

• AI-native design
• Spatial awareness
• User empowerment
• Simple interactions → emergent behaviors

---

License

MIT