Full-text Search for Verifiable Credential Metadata on Distributed Ledgers

TL;DR

This paper introduces a full-text search framework for credential metadata on Hyperledger Indy, enabling efficient discovery of credential types directly from blockchain data without relying on third-party sources.

Contribution

It presents a novel full-text search solution integrated with Hyperledger Indy, allowing users to efficiently find credential types using local ledger copies and a search engine.

Findings

Effective retrieval of credential types using full-text search.

Prototype implementation demonstrates feasibility and efficiency.

Reduces reliance on third-party credential information.

Abstract

Self-sovereign Identity (SSI) powered by distributed ledger technologies enables more flexible and faster digital identification workflows, while at the same time limiting the control and influence of central authorities. However, a global identity solution must be able to handle myriad credential types from millions of issuing organizations. As metadata about types of digital credentials is readable by everyone on the public permissioned ledger with Hyperledger Indy, anyone could find relevant and trusted credential types for their use cases by looking at the records on the blockchain. To this date, no efficient full-text search mechanism exists that would allow users to search for credential types in a simple and efficient fashion tightly integrated into their applications. In this work, we propose a full-text search framework based on the publicly available metadata on the Hyperledger Indy ledger for retrieving matching credential types. The proposed solution is able to find credential types based on textual input from the user by using a full-text search engine and maintaining a local copy of the ledger. Thus, we do not need to rely on information about credentials coming from a very large candidate pool of third parties we would need to trust, such as the website of a company displaying its own identifier and a list of issued credentials. We have also proven the feasiblity of the concept by implementing and evaluating a prototype of the full-text credential metadata search service.