It would be great to add a bunch of documentation:
- Design of each part of the system (from
TableOperations to Expressions, etc.)
- Tradeoffs taken in the design of the table format
- Context bounds - what is and isn't in scope for Iceberg
- What you need to do to replace Hive with your own metastore
- Bootstrapping your own use of Iceberg
- Reasoning for limiting what's possible with Avro schemas
It would be great to add a bunch of documentation:
TableOperationstoExpressions, etc.)