Apache iceberg does not solve the root of the problem. It slightly decouples a tightly coupled system (proprietary db implementations) but moves the coupling to a different layer, the metadata management. The problem that needs to be addressed is lock-in of the data stack. Sure, the data format itself is open, but that is kind of irrelevant when the catalog is what provides the utility. I still have to copy all my data if I want to switch catalogs. A positive to be recognized is the universal data types supported, but that was not the major problem IMO. What is the point of an open table format if you still get locked in to vendors or technologies via the catalog?