INFORMATION ARCHITECTURE AND NAVIGABILITY OF LARGE SCALE DIGITAL LIBRARIES
Keywords:
Information, Information Architecture, Navigability Large Scale Digital LibrariesAbstract
Background: Large-scale digital libraries aggregating petabyte-scale collections, such as Europeana with 50M+ multilingual artifacts and HathiTrust with 17M volumes, face critical usability challenges. Poor information architecture leads to unmanageable result sets, with up to 65% user abandonment when confronted with unfiltered million-hit results.
Purpose: This paper examines structural design principles that enable equitable discovery and navigability in large-scale digital libraries, with specific attention to Global South infrastructural constraints and the inclusion of indigenous knowledge systems.
Methodology: The study analyzes taxonomic evolution from rigid LCSH hierarchies to LOD ontologies, navigational paradigms including faceted browsing versus cognitive overload in heterogeneous metadata ecosystems, and AI-driven semantic navigation using BERT embeddings and knowledge graph traversals. It further evaluates emerging technologies such as predictive faceting, progressive disclosure, and server-side rendering for low-bandwidth contexts.
Findings: Faceted browsing resolves 73% of ambiguous queries, while progressive disclosure in DSpace reduced three-click abandonment by 72%. Server-side rendering achieved 92% task completion in 2G network environments where cloud-only interfaces failed. Hybrid human-AI architectures are needed to address metadata heterogeneity, with only 37% Dublin Core field completeness across national aggregations. Persistent challenges include format obsolescence and Western-centric controlled vocabularies that marginalize indigenous scholarship.
Conclusion: Information architecture is an existential determinant of large-scale digital library success. Librarians must evolve from catalogers to ontology engineers skilled in RDF/SPARQL curation and semantic stewardship. Decolonized standards, guided by IFLA and enabled through OAI-PMH interoperability, are essential to ensure navigational systems do not exclude the diverse voices that petabyte collections are intended to liberate.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.