Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Metadata Pipeline topic
No spam. Unsubscribe anytime.
Inside the Census Bureau's GPMS and automated pipeline for publishing geospatial metadata
Summary
Census staff described an internal GPMS schema, automated generation maintained by Decennial IT, XSLT-driven transformations to ISO outputs, and a multi-step validation flow; the Bureau said it creates and harvests over 33,000 metadata files each year and runs post-harvest QC.
Census geographers gave a detailed walkthrough of the agency's metadata life cycle. They described an internal Geospatial Product Metadata Standard (GPMS) used as a compact, standard-agnostic source record; GPMS records are programmatically transformed via XSLT into ISO-compliant outputs, enabling multiple public-facing formats without rebuilding source databases.
The team said the Bureau generates and harvests "over 33,000 metadata files each year," uses automated validation checks (inventory, ISO schema validation, invalid-character checks, XML difference analysis, URL checks), and runs a post-harvest QC using custom Python scripts against data.gov to ensure public portals match the internal harvest source.
