did:orb

The Orb DID Method is a decentralized identifier system built on a federated and replicated Verifiable Data Registry (VDR). Unlike blockchain-based DID methods, Orb operates through a network of interconnected servers that write, monitor, witness, and propagate DID operations without requiring a common blockchain for coordination.

The method uses a content-addressable architecture where DID operations are batched into graph nodes that can be discovered through both domain-based mechanisms and Distributed Hash Tables (DHTs). Each server in the network can act as a witness, maintaining an append-only Merkle Tree log of observed transactions.

Key features include:

• Self-certifying DIDs inherited from the Sidetree protocol
• Decentralized federation using ActivityPub for propagation
• Witness-based timestamping for resolving operation ordering
• Content-addressable storage for DID document batches
• Web API and DHT support for resource discovery

The architecture consists of three primary server roles:

1. Writers: Accept and batch DID operations
2. Witnesses: Provide timestamping and ledger services
3. Resolvers: Handle DID resolution requests

Technical Implementation:

DID Format:
did:orb:<anchor-hash>:<did-suffix>

Key Components:

• Uses ActivityStreams and ActivityPub for federation
• Implements RFC7033 for endpoint discovery
• Follows Sidetree protocol for DID operations
• Utilizes RFC6962 for witness ledgers

Security Features:

• Tamper-evident operation history through Merkle Trees
• Multi-witness validation for transaction timing
• Branch resolution rules for conflicting updates
• Content addressing for immutable storage

The method is particularly innovative in its:

1. Blockchain-free Architecture: Operates without requiring consensus
2. Flexible Federation: Allows organic network growth
3. Witness System: Provides relative timing without global consensus
4. Hybrid Discovery: Supports both Web APIs and DHTs

The technology stack includes:

• ActivityPub for federation
• Sidetree for DID operations
• Content-addressable Storage for data persistence
• WebFinger for service discovery
• Merkle Trees for witness ledgers