1 of 18

The Decentralized Archival Resource Key (dARK)

A decentralized, scalable solution for persistent identifiers in scholarly communication. Developed by Ibict with support from LA Referencia and SCOSS funding.

Lautaro Matas, LA Referencia�Washington Segundo, Ibict

2 of 18

The Inequity Challenge

Access to global research infrastructure is often unequal. Many institutions in the Global South face significant financial and technical barriers. This limits their participation in the global scholarly communication landscape, creating a digital divide.

PIDs are crucial for linking research outputs. However, current PID systems can exclude institutions with limited resources. We must develop solutions that promote true global inclusiveness.

3 of 18

Challenges in the Global South

Costs

Traditional PID systems impose financial barriers for institutions.

Centralized Models

Reliance on centralized systems limits accessibility and sustainability.

Limited Participation

Barriers hinder full engagement in global scholarly discourse.

4 of 18

Key Motivations

Need for Decentralization

Reducing dependency on centralized models

Research Assessment

Building robust research graphs and indicators

Global South Challenges

Addressing lack of persistent identifier coverage

ARK as Alternative

Viable, low-cost solution for local providers

5 of 18

Archival Resource Key (ARK)

Cost-Effective

Organizations can generate unlimited identifiers at no cost.

Flexible

Designed for reliable referencing of information objects.

Greater Control

Institutions maintain control over their data and metadata.

https://arks.org

6 of 18

Long-term Objectives

Open Infrastructure

Non-centralized system for unique persistent identifiers accessible to all

Resolution Services

Decentralized resolution service interoperable with other PID services

Metadata Preservation

Decentralized preservation of metadata for consistent research graphs

7 of 18

Introducing dARK

What is dARK?

A decentralized, blockchain-based implementation of the ARK Persistent Identifier standard.

Why ARK?

Open, flexible, and cost-free to use; well-suited for institutions with limited resources.

Built as a Public Good

Supported by IBICT (Brazil) and LA Referencia, with funding from SCOSS.

Core Principles

Openness, decentralization, interoperability, and long-term preservation of research assets.

dARK leverages the open ARK standard with blockchain technology.

This creates a robust, community-governed PID system. It prioritizes accessibility and long-term viability for global research.

8 of 18

Why Decentralization Matters

Empowers Local Institutions

Organizations can create, manage, and resolve their own PIDs, reducing dependency on external agencies.

Reduces Financial Barriers

No per-identifier or annual fees; shared infrastructure dramatically lowers operational costs.

Ensures Digital Sovereignty

Control over metadata and research assets stays with national and institutional actors.

Increases Resilience and Sustainability

A distributed system prevents service disruption even if some nodes fail or exit the network.

Fosters Open and Transparent Governance

Participating institutions co-govern the system, encouraging equitable and sustainable practices.

Decentralization is key to a more equitable research ecosystem. It puts control and resources directly into the hands of local institutions. This fosters self-reliance and sustainability.

9 of 18

Technical Implementation

Blockchain Consortium Network

Nodes hosted by institutions collectively store and control data.

Commodity Hardware

Institutions can set up nodes using existing platforms.

Abstraction Layer (API)

Handles metadata transfer, authentication, and system interactions.

Global Resolver Integration

Enhances discoverability through n2t.info resolver.

10 of 18

System Architecture Overview

Service Layer

Interfaces with users and external systems through specialized components

Core Layer

Built on permissioned blockchain with Proof of Authority consensus

11 of 18

Service Layer Components

dARK Resolver

Enables persistent identifier resolution through nt2.info system

dARK Minter

Creates and registers new PIDs in the system

dARK Dashboard

Provides monitoring and administrative capabilities

dARK API

Facilitates communication with the underlying blockchain

12 of 18

First implementation: Oasisbr Aggregator (Brazil)

NAAN Assignment

ARK Alliance assigns prefixes to institutions

Mass Mintings

System for bulk PID creation implemented

Metadata Enrichment

Metadata update with ARK pid and shared with the repository

Global Resolution

Integration with n2t.info enhances discovery

13 of 18

Brazilian Open Science Ecosystem

5.5M+

Digital Objects

Aggregated in the Oasisbr portal

1,600

Open Access Sources

Contributing content to the platform

125

Digital Repositories

Brazilian institutions participating

600K

dARK PIDs Assigned

During the pilot phase

14 of 18

Next Step: Expanding Across Latin America

Building on the success in Brazil, dARK is ready for broader adoption. LA Referencia will facilitate expansion to other Latin American countries. This will further enhance regional research infrastructure. The goal is to create a stronger, more connected research community.

Training and support will be provided to new participating institutions. This ensures smooth integration and maximizes the benefits of decentralized PIDs. Collaboration is key to this expansion.

15 of 18

Expanding and Innovating Federated Open Science Infrastructure in Latin America (2026–2028)�IOI / LA Referencia

A three-year initiative led by LA Referencia to strengthen and expand the regional open science infrastructure with support from the IOI Fund. This project builds a more equitable, interoperable and sustainable ecosystem for research outputs across Latin America.

Component A

Expansion of the aggregation platform + AI/LLM services (semantic multilingual search, enrichment)

Component B

Deployment and scaling of dARK, a decentralized ARK PID service with an open plugin model for repositories

Component C

Regional data repository for orphan/underserved datasets + a federated support program for institutional platforms

All outputs released as open-source, aligned with FAIR, COAR, OpenAIRE and international best practices. The initiative focuses on regional sovereignty, long-term sustainability, and inclusion of all Latin American countries.

16 of 18

Component A – Strengthening Aggregation and Enabling Multilingual Discovery

Infrastructure Expansion

Expansion of harvesting and aggregation infrastructure to 10 new national nodes by 2028, with full alignment with OpenAIRE Guidelines, COAR vocabularies, and FAIR practices for metadata quality and interoperability.

Multilingual Semantic Search

Development of a multilingual semantic search engine using open LLMs for meaning-based discovery (ES–PT–EN and beyond)

AI-Powered Enrichment Pipeline

  • Multilingual keywords and summaries
  • Subject classification
  • Affiliation normalization using PIDs (e.g., ROR)

Seamless Integration

Integration with existing platforms and national ecosystems through open APIs

17 of 18

Component B – dARK: Decentralized ARKs for Resilient, Public Good, and Sovereign PIDs

dARK implements ARK persistent identifiers over a permissioned Hyperledger Besu blockchain with IPFS for distributed metadata preservation. This provides a regional PID service as public-good, auditable, and interoperable with existing repository workflows.

ARK Minting and Resolution

Core capability for creating and resolving persistent identifiers

Replicated IPFS Storage

Distributed storage of metadata packages for preservation

Dashboard and Governance Tools

Management tools for PID authorities

Plugin Architecture

Lightweight connectors for platforms to issue ARKs through dARK

Scaling from the Brazilian pilot to at least 10-country federated network by 2028.

18 of 18

Component C – Regional Dataverse Data Services and Capacity Building for Latin America

Creation of a regional data repository using Dataverse to support researchers without institutional data services, long-tail and orphan datasets, and regional programs requiring shared stewardship.

1

2026

Alpha version + helpdesk and training resources in ES/PT

2

2027

Controlled beta and pilots

3

2028

Full service with shared governance and sustainability model

Federated support program to assist institutions deploying or improving Dataverse. Strengthens regional capacity through documentation, training, onboarding, and technical assistance, contributing to a more inclusive and resilient research data ecosystem across Latin America.