Joe Celko's Data and Databases: Concepts in Practice (Morgan Kaufman Series in Data Management Systems) [Englisch] [Taschenbuch]

If you need an introductory book on data and databases then DON'T buy this book, as it's nothing of the sort. If you need a book giving a complete, thorough grounding in all aspects of relation theory, data and databases, again DON'T buy this book. However, if you work with databases for a living, already have a reasonable understanding about them, and want an assorted collection of discussions about the nature of data and databases then certainly DO consider buying it - you may find it gives deeper insights in this case.

The title is interesting - "Data and Databases", which reflects the book itself. The first half of the book mostly discusses the nature of DATA, with the second half mostly about DATABASES. "Concepts" - certainly, the book is mostly about concepts. "In Practice" - definitely NOT the case, as the book is mostly conceptual with very little of a direct practical nature.

I found the chapter on Keys most useful - an in depth discussion about surrogate keys, which most books barely mention at all. However, even this chapter reflects the general nature of the book. For any given topic, Celko picks a specific aspect which interests him, discusses it at length, but gives very superficial coverage of other equally important aspects of that topic.

The main problem with this book is that the description on the back cover is a blatant misrepresentation of its contents. It leads you to believe it is a comprehensive grounding in all aspects of data and databases, but it simply isn't. Celko admits as much in the Preface, stating it is "a collection of ideas...not a complete, formal text".

Previous reviews range from 1 star to 5 stars, but both types of reviews are actually equally valid depending on where the reader is coming from. If you are new to databases and looking for an introductory text then this book is NOT for you, despite the publishers claiming so on the back cover. If you are experienced with databases then certainly DO consider it, as it does give insights you don't normally come across.

I would have given four stars, but can only give 3 due to (a) the disjointed presentation of topics, and (b) the blatant misrepresentation by the publishers about the book's contents. However, I do recommend it to experienced database professionals.

(p.s. - before someone else goes on about reviewers who don't what they are talking about, I've spent over a decade earning my living developing databases, and spent literally thousands of hours doing formal study to Masters level, so I do have a reasonable idea what I'm on about).

There is one thing I do not like that much in his books. His likes to show that he knows a lot or knows where to find it, without any use for the book. This irritates me a bit. For example, why on earth list the axioms of intuitionist mathematics. I suppose I am one of the few readers who heard about intuitionism before and it is of certainly no help in this book. It is not there for the purposes of the book! Or another example, section 1.2.2 tells a bit about bad math. He tries to show that reporters cannot do simple math. But why does he assume there is a linear relation between weight and burned calories? May be there is a fixed amount of calories that you always burn, no matter what you are doing. I am not an expert on calories, but his logic of showing somebody's errors is not always correct. The correct answer for this calorie problem should be: we do not know and the 'proof' of the reporter is wrong. This does not mean the proposition is wrong! Another mathematical error, he writes that: "The idea of a limit is that there is a value that the sum never exceeds". Well, this is the definition of a upperbound. In case the function is non-decreasing the smalles upperbound will be the limit, it is easy to proof this. Of course, this is not a math book, but if you use the techniques you should be correct.

However, still a nice book to have and Joe mailed me always back when I had a remark or question. This care deservers one extra star!