> i meant no two tables under the same catalog can have the same location
This is a stricter requirement than we have for Iceberg tables. Are we really going to enforce this? How will we do it efficiently? If not, let's not put it in the spec. > we do not have any update support It would be trivial to add update support for generic entities. Why canonicalize this restriction in the spec? We don't, for example, currently detail a restriction around the fact that you can't change a generic table's format. > generic tables are designed for non-Iceberg tables today, I don't actually think this is true. There's nothing about generic tables that make them more useful for Delta tables than Iceberg tables, for example. On the contrary, I initially proposed the name "generic" in part to capture that generic tables are a catch-all type not specific to any format (including Iceberg). More practically, GenericTableEntity is the type I'm most likely to look to for the conversion service, which means it will indeed be used to represent Iceberg tables. > The multi-location support in Polaris seems not very well also, the overlap check and credential vending seems all done with one location This is not true.
