Archives, catalogs, and research databases are some of the most rewarding sites to build and some of the worst to maintain badly. They hold thousands of records, get edited by several people over years, and need to be browsable by date, place, person, category, and every other facet a researcher might come at it from. Done the usual way, that means a heavy database, a server to secure and patch, and a hosting bill the project can’t sustain after the grant runs out.
A database without the database
I build these as structured content: every record is a file, edited through a friendly CMS (Sveltia / Astro), with no traditional database to host, secure, or break. That sounds like a technicality, but it’s the whole point. It’s what makes a large archive fast, durable, and affordable to keep running long after launch.
- Thousands of records, organized properly, modeled around how researchers actually navigate the material rather than forced into a generic template.
- Faceted search and filtering across the full dataset, so a visitor can narrow by any dimension that matters.
- Multi-user editing for projects with several contributors, without everyone needing to be technical.
- No fragile infrastructure. Nothing to get hacked, nothing to fall over, nothing that demands a developer on retainer just to stay alive.
Selected work
The clearest example is Shakespeare in Spain, an open-access catalogue for York University and the Universitat de Barcelona: 681 catalogued copies traced across 84 Iberian institutions, plus 389 dramatic works and 141 authors, edited by multiple contributors and browsable through seven stacked facets, with no database to maintain. I’ve built and contributed to other large historical databases too, including the GEES Project, with another historical archive in progress.
Who this is for
Universities and research projects, historians and archivists, cultural organizations, and anyone sitting on a large, structured body of material that deserves to be properly browsable and needs to stay online and affordable for the long haul, not just until the launch budget runs out.