A lot of infrastructure work is needed to make the SQL experience seamless work for unstructured data. And at the most part we do fork the open core data warehouse and build on top of it.