Web2 provides a complete identity system at the cost of excessive centralization, where users have no control over their own data. The emergence of Decentralized Identifiers (DIDs) meets this demand by shifting user identity provision from centralized platforms to a "decentralized" approach. Unlike assets, identity holds a social nature. If a person lives alone on a deserted island since childhood, names would be meaningless. This social aspect requires identity to be somewhat public, which contradicts the inherent principles of blockchain and Web3's "application primitives," resulting in various challenges. Existing Issues

Currently, our identity schemes mostly involve attaching identifiers to decentralized platforms, including domain services, content services, asset services, and more. This mechanism leads to two outcomes:

  1. Multiple platforms maintain user identities to cater to different scenarios and needs. Users are required to use multiple identity platforms in their daily lives, which complicates their control over their identity and data.
  2. Eventually, only a few platforms (2 or 3) achieve monopolistic status, leaving the majority of users restricted to these platforms. Despite being decentralized platforms, this situation is still centralized.

These two outcomes reveal a core paradox of DID products: when you start implementing DIDs, it is no longer a DID (this paradox is reflected in the product by introducing an additional ID to control all the user's other IDs, forming a DID in a sequential manner—for example, mapping different addresses to ENS or establishing a DID account through ceramics and adding other addresses to the controller). DID with a bidirectional linked list structure

We aim to establish a DID that satisfies the following two conditions:

Overview

By associating user IDs in a manner similar to a bidirectional linked list, we enable any third party to access the predecessor and successor nodes from any ID and obtain complete identity information. In the field of computer science, this is a fundamental data structure that proves valuable in the context of DIDs.

Each user's ID represents a fragment of their life, and combining all these fragments forms the complete user identity. This is akin to the montage technique used in movies. We name it the Montage Protocol. Currently, this is merely a concept and has not been fully implemented. It has several important core conventions:

  1. Discoverability: User interaction with the protocol should be public and discoverable, such as events on the blockchain.
  2. Neutrality: It should be independent of any platform, including decentralized ones.
  3. Consistency: The results of user interactions with the protocol should be consistent.
  4. Uniqueness: User identity should be unique globally.

These characteristics make it a genuinely decentralized identity. Ecological Service Providers

Service providers adhere to the protocol and monitor user actions across all deterministic platforms to form a consistent identity configuration. This distinguishes it from RSS3-like products, where RSS3 still consolidates all user data into a single platform, requiring direct user interaction with RSS3 itself. The Montage Protocol has no admission system, allowing any service provider to obtain the complete user identity.

Wallets

In Montage, user identities are public, which may initially seem unfavorable to privacy protection. However, the Montage Protocol is highly beneficial for privacy. Users often have multiple IDs within a wallet, and using different addresses interchangeably can easily lead to privacy leaks. Wallets can utilize the Montage Protocol to collect users' public identity information and group identities within the wallet, making it easier for users to associate usage scenarios with specific accounts at the interaction level.