RFC: Community Issuance Network — trusted institutions as identity entry points

[imajin-jin]

📄 Canonical document: docs/rfcs/RFC-14 — this discussion is for community input; the RFC source of truth lives in the repo.

---

Context

Greg Mulholland's "Commons Layer" proposal (March 2026) argues that Imajin's identity architecture has implications beyond the product — it's a prototype for public digital identity infrastructure. The key insight: any trusted institution can be an identity issuance point.

Related: #271 (Progressive Trust Model), #247 (Cultural DID), #249 (Plugin Architecture), Discussion #269 (MJN Token Economics / Foundation)

The Insight

Imajin's EventDID check-in model is, structurally, a community-anchored identity issuance network. A person scans a ticket, passes an ID check, receives a DID. The physical body is the proof of work.

This generalizes. The gate is any verifiable in-person interaction at a trusted institution:

Institution	Issuance Context
Imajin events	EventDID check-in (exists today)
Libraries	Library card issuance → DID
Credit unions	Account opening → DID
Community orgs	Membership verification → DID
Medical practices	Patient verification → DID
Schools/universities	Student verification → DID

Each institution becomes an issuance point on the network. The DID they issue carries an attestation: "verified in-person by [institution DID] on [date]."

How This Maps to Existing Architecture

What already exists

• Ed25519 keypair identity — no platform issues it, no platform can revoke it
• EventDID — events as first-class identity-issuing entities
• Federated nodes — anyone can run a node (registry.imajin.ai)
• MJN Foundation (planned) — protocol stewardship separate from Imajin Inc.

What this adds

• Institutional DID type — a new DID category for trusted issuance points (or an Org DID with issuance attestation rights)
• Issuance attestations — "DID X was verified in-person by Institution Y" as a first-class attestation type
• Issuance point registry — how institutions register as trusted issuers (probably through the federated node model)
• Cross-institution trust — a DID issued at a library carries weight at a credit union because both are on the same trust network

Imajin Product vs. Imajin Protocol

Greg's framing (and Ryan's existing architecture) separates these cleanly:

	Imajin Protocol (Foundation)	Imajin Product (Inc.)
Owns	DID spec, trust primitives, settlement protocol	Reference implementation, UX, community tools
Governed by	Foundation board, weighted by community contribution	Imajin Inc. (Canadian corp)
Revenue	Protocol fees (mint/burn spread, settlement micro-fees)	Operator excellence, premium features, node hosting
Analogy	HTML/HTTP	Netscape/Chrome

The protocol is a public good. The product competes on operator excellence. Open protocol makes the product more valuable because the network is larger.

Why This Matters

There is currently no public option for digital identity. Every major system (Google, Apple, Meta) extracts value from the identity it holds hostage. Government IDs have no digital native form in most jurisdictions.

A community-anchored issuance network means:

• Identity doesn't require a smartphone with a proprietary OS
• Identity doesn't require an advertising account
• Identity doesn't require a government to issue it
• Identity requires a trusted human to verify that a body exists — and cryptographically record that verification

Open Questions

1. How do institutions register as issuance points? Through the federated node model, or a separate registration path?
2. What level of verification is required? Government ID check? Or is institutional trust sufficient (library card, membership roll)?
3. How does this interact with the MJN Foundation governance? Does the Foundation certify issuance points?
4. Timeline: is this Year 1 (product), Year 2 (protocol), or Year 3 (public infrastructure)?

Credit

Core concept from Greg Mulholland's "The Commons Layer" (March 2026). Architectural grounding from Imajin's existing federated node model and MJN Foundation plans.