from the blog: "This is a very interesting new development, making DuckDB potentially a suitable replacement for lakehouse formats such as Iceberg or Delta lake for medium scale data."

I don't think we'll ever see this, honestly.

excellent podcast episode with Joe Reis - I've also never understood this whole idea of "just use Spark" or you gotta get on Redshift.