How catalog federation gives Iceberg tables in several regions one authorization model and audit point, what it cannot fix, and how to build the topology.
Federate first so analytics work now, migrate what benefits from migrating, and leave the rest where it is indefinitely.
Parquet files have a version field that doesn't reliably signal feature requirements. A new versioning discipline is coming, borrowing from Iceberg's.
A large share of production transformations fit comfortably on one machine. PyIceberg, DuckDB, and branch isolation give you a production path that debugs.
