Abstract
Conventional online social networks (OSNs) are implemented in a centralized manner. Although centralization is a convenient way for implementing OSNs, it has several well known drawbacks. Chief among them are the risks they pose to the security and privacy of the information maintained by the OSN; and the loss of control over the information contributed by individual members. These concerns prompted several attempts to create decentralized OSNs, or DOSNs. The basic idea underlying these attempts, is that each member of a social network keeps its data under its own control, instead of surrendering it to a central host; providing access to it to other members of the OSN according to its own access-control policy. Unfortunately all existing DOSN projects have a very serious limitation. Namely, they are unable to subject the membership of a DOSN, and the interaction between its members, to any global policy. We adopt the decentralization idea underlying DOSNs, complementing it with a means for specifying and enforcing a wide range of policies over the membership of a social community, and over the interaction between its disparate distributed members. And we do so in a scalable fashion.
Introduction
The dominance of centralized Online Social Networks (OSNs) like Facebook and X stems from their ease of management and ability to enforce global interaction rules. However, this centralization creates significant privacy risks and "honeypots" for data breaches. While Decentralized OSNs (DOSNs) offer a solution by allowing users to retain data ownership, they often lack the governance necessary for structured communities. Without a central authority, enforcing global policies—such as membership verification or role-based access—becomes a significant challenge. This paper introduces the Online Social Community (OSC) model, which leverages Law-Governed Interaction (LGI) to provide centralized-style governance within a decentralized architecture.
The OSC Model and Law-Governed Interaction
The core of the proposed solution is the OSC-model, which treats a social network as a formal community governed by a digital "constitution" or law. This is achieved through Law-Governed Interaction (LGI) middleware.
LGI Middleware and Dual Control
Unlike traditional systems where access control is typically enforced only at the receiver's end, LGI employs a "dual control" mechanism. Every member of the community is associated with a local Controller, a trusted software mediator. When a member (agent $x$) sends a message, it is first intercepted by their local controller ($T_x^L$), which evaluates it against the community law ($L$). If compliant, the message is signed and sent. Upon arrival, the receiver's controller ($T_{x'}^{L'}$) verifies the signature and the law's compliance before delivering it to agent $x'$.
Community Architecture
A decentralized community consists of disparate members and optional support components. Members maintain their own data (DB) and interact through their controllers. Support components, such as a Directory for member discovery or a Certificate Authority (CA) for identity verification, assist the network without acting as a central point of failure for policy enforcement.
Key Insights
- Decentralized Governance: The OSC model proves that a central server is not required for central policy. If every local controller follows the same immutable law, the system achieves global consistency.
- L-Trust (Law-based Trust): Controllers use cryptographic handshakes to ensure they are running the same version of the law. This prevents users from bypassing policies by tampering with their own local clients.
- Stateful Policy Enforcement: Laws in LGI are not merely static rules; they are stateful. Controllers can track interaction history, allowing for complex behaviors like revoking access based on specific sequence of events (e.g., a "fire" message from a manager).
- Performance: The computational overhead of local enforcement is minimal, measured at approximately 50 microseconds per message, which is negligible compared to standard network latency.
Implementation and Results
The authors validated the OSC model through a proof-of-concept implementation of an enterprise micro-blogging community ($B_E$) with over 200 users. The system successfully enforced:
- Membership Control: Utilizing digital certificates for joining and enabling administrative roles to revoke memberships.
- Role-Based Interaction: Restricting specific communication types (e.g., official announcements) to authorized roles.
- Subscription Filtering: Automating the "Follow" mechanism where publishers' controllers filter subscribers based on organizational attributes.
Conclusion and Future Work
While the OSC model successfully decentralizes policy enforcement, it faces challenges in global data analysis (e.g., calculating network-wide metrics) and search security. Future work aims to integrate LGI with Distributed Hash Tables (DHTs) to secure decentralized search and explore "conformance hierarchies" to allow sub-groups to define specialized rules within a broader community framework.
Summary updated