This is the first in a multi-part series comparing the atproto and Secure Scuttlebutt (SSB) protocols, with the ultimate goal of exploring how to bridge the two. In this post, I will provide a high-level overview of the foundational concepts and philosophies that underpin each protocol.
AT Protocol: Federated and Composable
The AT Protocol (atproto) is designed for "usable decentralized social media." It aims to provide a user experience that is familiar to users of centralized social media platforms, while still offering the benefits of decentralization, such as data portability and censorship resistance.
At its core, atproto is a federated protocol. User data is stored in Personal Data Servers (PDSes), and a network of Relays crawls this data to create a global "firehose" of public activity. This architecture allows for large-scale applications and efficient data discovery, but it also introduces a more centralized element into the system.
Identity in atproto is managed through a combination of mutable handles (like usernames) and immutable Decentralized Identifiers (DIDs). This provides both a user-friendly way to identify users and a secure, cryptographic foundation for the network.
Secure Scuttlebutt: Radically Decentralized and Peer-to-Peer
Secure Scuttlebutt (SSB) takes a more radical approach to decentralization. It is a peer-to-peer protocol where data is shared through a gossip network. There are no central servers or relays; instead, data propagates through the social graph as users connect to each other directly or through "pubs" (public peers).
Identity in SSB is based on a user's public key, and trust is established through a "web-of-trust" model. Users follow each other to replicate their append-only logs, creating a subjective view of the network based on their social connections.
This architecture makes SSB highly resilient and censorship-resistant, but it also presents challenges for data discovery and scalability.
Key Philosophical Differences
The fundamental difference between atproto and SSB lies in their design trade-offs:
- atproto prioritizes usability and scalability by incorporating a federated layer for data indexing and discovery.
- SSB prioritizes radical decentralization and censorship resistance by relying on a pure peer-to-peer architecture.
These differing philosophies have profound implications for how data is represented, how identity is managed, and how applications are built on each protocol. In the next part of this series, I will delve into the technical details of data representation and identity in both atproto and SSB.
