Skip to content

Add dogfood file explaining dogfooding for the project #55

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
oldsj wants to merge 1 commit into from
Closed

Add dogfood file explaining dogfooding for the project #55

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 26 additions & 0 deletions dogfood
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
# Dogfooding Mainloop

**Dogfooding** for mainloop means using mainloop itself to develop mainloop. It's the practice of eating your own dog food—using the product you're building in the same way your users will.

## What It Means for Mainloop

Mainloop is an attention management system that uses spawned AI sessions to handle background work while you stay focused in the main conversation. Dogfooding mainloop means:

1. **Using spawn_session for development tasks** - Instead of manually handling feature development, bug investigation, or refactoring in a single conversation, spawn dedicated sessions to parallelize the work.

2. **Keeping the main thread focused** - The main conversation thread stays clean and focused on decisions, planning, and integration. Details, iteration, and implementation live in spawned sessions.

3. **Leveraging notifications and task queuing** - Sessions surface results back to the main thread as notifications, maintaining a unified inbox of what needs attention.

4. **Testing the UX in real scenarios** - By using mainloop to build mainloop, the team experiences firsthand where friction points exist, what the notifications feel like, and how the session workflow scales.

## Practical Examples

- **Feature development**: Spawn a session to implement a new feature, investigate design patterns, run tests, and prepare a PR—all in parallel with other work.
- **Bug investigation**: Create a session to dig into logs, trace code paths, and propose a fix while the main thread stays available.
- **Documentation updates**: Spawn a session to update docs, examples, or architecture diagrams without blocking the main conversation.
- **Code review cycles**: Use sessions to address feedback, run CI checks, and iterate—the main thread only gets pinged when ready for final review.

## Why It Matters

Dogfooding validates that mainloop actually solves the problem it was built for: **keeping humans focused while parallel AI work happens in the background**. If building mainloop itself becomes easier and more manageable with mainloop, then the product works.
Loading