I’m researching validation practices in large-scale STAC ingestion pipelines.
I recently confirmed a public case where STAC projection metadata was structurally valid, but inconsistent with the actual GeoTIFF header.
In a system at Planetary Computer’s scale, are metadata–asset inconsistencies like this something you encounter often enough to validate systematically?
If so, where are they typically caught today — during source onboarding, ingestion, asset processing, or post-publication QA?
I’m mainly trying to understand whether asset-aware validation is a recurring ingestion concern, rather than a one-off producer bug.
I’m researching validation practices in large-scale STAC ingestion pipelines.
I recently confirmed a public case where STAC projection metadata was structurally valid, but inconsistent with the actual GeoTIFF header.
In a system at Planetary Computer’s scale, are metadata–asset inconsistencies like this something you encounter often enough to validate systematically?
If so, where are they typically caught today — during source onboarding, ingestion, asset processing, or post-publication QA?
I’m mainly trying to understand whether asset-aware validation is a recurring ingestion concern, rather than a one-off producer bug.