I owe this post to Mr. Lex de Haan with whom I was bantering when he recommended the book. The book should be entitled "An Introduction to Relational Theory for an obnoxious know-it-all Oracle DBA". Book is written in a very clear and easy to understand fashion. It gives a general overview of the theory and states the principles, rules and goals. That is all very nice. My only objection to the book is that it failed to establish connection to the strictly mathematical foundations of the theory. Relation is, strictly speaking, a subset of Cartesian product - any subset is a relation. Mathematics knows many relation types: symmetric,transitive,antisymmetric, relation of ordering and alike. None of them are mentioned. Union and intersect are set theory operations. None of the set theory was mentioned in its strict, formal form which I find unacceptable when explaining a theory that is essentially a part of mathematical set theory. Axiom of choice, Zorn's lemma and well-ordering theorem are not necessary to mention and explain in a book for a "database professional" but, in my opinion, the author did shy away too much from using mathematics. Cary Millsap did not make such mistake. I don't want to keep this esteemed audience under suspense any longer: yes, I am a mathematician, with college in mathematics. The book is good, but it needs more math. -- Mladen Gogala Oracle DBA Ext. 121 -- //www.freelists.org/webpage/oracle-l