Strategic Direction: Should Reeve exist alongside OpenClaw?

[reubenjohn]

The Question

Reeve and OpenClaw take fundamentally different approaches to the same problem. Before investing more effort, I'd like community input on whether this architectural bet makes sense.

The Core Trade-off

	OpenClaw	Reeve
Paradigm	Custom runtime	Orchestrator wrapping CLIs
Session	Continuous context	Isolated per wake-up
Memory	In-session read/write	Git-versioned Desk
Observability	Debug via logs	Glass Box: edit the brain

Four Options

1. Abandon - OpenClaw covers this space well enough
2. Merge - Contribute Reeve's ideas (Pulse Queue, Desk pattern) to OpenClaw
3. Compete - Build Reeve as a full alternative
4. Coexist - Reeve as specialized orchestrator, OpenClaw as runtime

What I'd Love to Hear

• Does the "session isolation" philosophy resonate with you?
• Is the "Glass Box" (editable Markdown brain) valuable?
• Would you use a Pulse Queue for proactive scheduling?
• Any other architectural feedback?

See the full comparison for details.

---

Note: This is a side project with limited maintenance bandwidth. Feedback helps prioritize.