Rethinking Data Discovery for Libraries and Digital Humanities

Harvard Library Innovation Lab built a serverless data discovery platform for their 18TB Data.gov Archive using client-side technologies. By leveraging DuckDB-Wasm, Parquet files, and HTTP range requests, they created a search interface that runs entirely in the browser while storing data on static hosting ($1/month). This approach eliminates traditional server costs and maintenance overhead while providing robust browsing, filtering, and search capabilities. The solution addresses a longstanding challenge for libraries and digital humanities projects: balancing rich data discovery features with sustainable, low-maintenance infrastructure.