Glossary
Search Index
A search index is a separate data structure, built from an underlying data source and optimized for fast querying, rather than the source data itself. Because it's a derived copy, it doesn't automatically track every change made to the source afterward — including changes to who's allowed to see a given piece of data, which the index has no inherent mechanism to notice or reflect.
This makes an index a specific, recurring blind spot: content can be findable through search well after the same content's access has been revoked at the source, simply because nothing re-synchronized the index's own copy or its own access checks to match. Keeping the two in sync is a design responsibility, not something that happens automatically by virtue of the index having been built from the source in the first place.