Rating: 
Summary: A Mixed Review
Review: When I saw that Jeffrey Ullman had a new Database text out I was excited, since my work focus has now moved into that area. I have read and learned from many of Ullman's books, such as the famous Dragon (compiler design) book, as well as the "White" (automata) book and his two texts on computer algorithms. These are classics and should be on the bookshelf of everyone who calls him/herself a computer scientist.This book is, however, a little disappointing. Most of it is good, some of it very good. But I do find some flaws in it. One of the glaring flaws deals with the attempt to extend the relational model from sets to bags (basically, to allow for duplicate tuples in relations.) This is the best attempt I've seen at formalizing "bag theory", but it introduces problems (some minor, others very serious) that aren't mentioned in the text. This review is too short and not the right place to expound on these problems. Chris Date's database text goes into most of them in substantial detail. In summary, this book is good, with many good examples. I find it very readable. But it is not as good as Chris Date's Intro to Database Systems for the serious database professional. Ullman's book is good for showing another perspective to Date's solid (but somewhat opinionated) treatment of relational database theory.
| 
