Maybe it's just the informal writing style and concentration on deficiencies in current implementations of SQL, but the whole things reads as if the author gave only a cursory glance at any relational model theory.