Customer Reviews

Relational Database Design Clearly Explained

5 star:		(11)
4 star:		(4)
3 star:		(3)
2 star:		(3)
1 star:		(2)

 

 

I've been programming for quite a while, but up until a couple of years ago, I had never done anything with databases. When I switched jobs and needed to learn, I asked a few friends where to begin. At the time they recommended "SQL For Dummies" and "Oracle8: The Complete Reference" -- starting with the first and then diving in to the second.

This didn't work so well for me. SQL For Dummies is actually a reasonable review of constructing basic SQL statements, but it didn't provide me with any conceptual framework for thinking about databases. The complete Oracle8 reference book certainly seemed to have a lot of material in it, but it was a bit too daunting (1300 pages) for a tutorial. I learned the basics that I needed to learn and have gradually assimilated things since then.

Recently I came across this book at Amazon and found that it had pretty good customer reviews, so I thought I'd check it out. It is excellent. This is the book I wish I had had from day one.

The book begins by reviewing basic concepts of databases and database design, plus by going over the various documents and diagrams that typically go along with databases. Then it briefly reviews the SQL one would use to create databases. And it ends with three detailed database design examples. These examples are pretty substantial -- in each case I read the description and thought "this is too complex a database to cover here" but the author broke the problem in comprehensible pieces, drew entity-relationship diagrams, and worked through the design.

I highly recommend this book to people just starting out with databases. You will probably need to follow it with something that teaches you more SQL, but that should follow, not preceed, an introduction like this.

The author claims to have condensed a college course down into a single 300 page book, which I would have trouble believing except that she apparently skipped over 1/2 of the course content. I do not think that this is a good book to explain Relational Databases, OR Database Design. The author gives a cursory explanation of each, then dives into an example DB with many hard-to-understand tables.

After confusing the reader with a 200,000 foot overview of Databases, the author goes into a very detailed explanation of the normalization process, which is IMPOSSIBLE to understand, and a long, drawn-out review of the theory of how a Relational Database Server should work, with no explanation of how current products do or do not adhere to this theory.

The author spends quite a bit of time plugging a Macintosh-Based ER system, all the way down to a section explaining the drawing environment of that *piece* of software.

Overall, this book was a waste of time. Buy Database Design for Mere Mortals by Michael J. Hernandez , a MUCH better book.

Jan Harrington accomplishes her task neatly, to clearly explain relational database design. I was very pleased with how quickly I was able to grasp fundamental concepts and I would recommend this book to anyone getting started with databases. I had hoped that the book would be perfect and would clearly explain every relevant concept but that was not the case. When it came to the three interesting case studies that concluded the book, the author used concepts that were never explained - control-break layout, parent entities, ISAM file organization, repeating groups, reblocking files. While not understanding these concepts did not stop me from grasping the fundamentals of database design it was frustrating and made clear that this is an introductory text and not the last book to read on the subject. There were also about a dozen typos but these were disconcerting rather than misleading.

In a book named "Clearly Explained" , you'd expect to have a minimum of confusion. That is not the case with this book.

Page(25) for example:

"Other many-to-many relationships include that between a child and her biological mother. A woman may have zero, one or more biological daugthers; a daugther has only one biological daughter"

Please correct me if I'm wrong but that does not describe a many-to-many relationship. There are other confusing examples in the book (pg 33 for example).

I found myself reading sections over and over because the examples were not that clear. Coupled with the errors, I found this book to be less than helpful.

It is typical in DB texts to error in the examples, such as to assume all phone numbers only need 7 INTs (we all live in the same area code, and country, right?) However, in this book there are so many serious inaccuracies that I found myself doubting all of the information presented. A quick example, pp4 on page 24 and again on pp2 page 25, on "one-to-many" relationships, "Other many-to-many relationships include that between a child and her biological mother." Uh, is she talking multiple embryonic transplants?

The information that *is* explained correctly is explained in a baffling, overly academic manner. For example, in explaining one-to-one relationships, "If we have two instances of two entities (A and B) called Ai and Bi, then a one-to-one relationship exists if at all times Ai is related to no instances of entity B and Bi is related to no instances of entity A or one instance of entity A." Right... way to painfully obfuscate a self-explanatory concept.

The author makes assertions, promising to back them up further in the book, and then never does so. She goes on to build additional conceptual elements based on these unsubstantiated assertions. You'll find yourself both distrusting the assertions, due to the numerous errors, but required to "believe" just to get to the next concept.

All around awful attempt.

I have to give this book a very mixed review. As many reviewers pointed out, there were some errors at key points, which muddied things up for someone new to databases. There were a few places where pages were added with little value; I really didn't need to see 6 pages of the "Mighy-Mite motors product catalog", the 20 pages of SQL CREATE TABLE statements, or even the chapter on CASE tools.

On the positive side, I appreciated the discussion of normalization, as well as "Codd's Rules". For those who have used keys packed with information, the section on "Avoiding meaningful primary keys" was useful. I appreciated the cursory mention of data flow diagrams; as I often think that database designers and database application developers would benefit from using these techniques to communicate the intersection of tables and process. I would have appreciated including this in the case studies. The case studies were thought provoking, and brought home some of the problems about many-to-many relationships.

My background is in non-relational databases and supporting them. This book explains the relational database theory in such simple, clear terms, that it was easy to relate the non-relational database that I'm familiar with relational database theory. This book was very objective about explaining the fundamentals of relational database theory without a lot of mis-directed 'fluff' or 'clutter' to boost the page count. I felt that the examples were great too.

With all due respect to the efforts of the authors, editors, and publishers, the book did have some typographical errors in key places which was very confusing. There were only a handfull of them. I'd suggest if you find yourself TOO confused (especially when reading about one-to-one, one-to-many, and many-to-many relationships), it's a clue that you just ran across a typo.

The book accomplished its main object: introduce the reader to database concepts without going into depth into any of the areas. This book should also be marketed as "Database for Dummies". I recommend this book to anyone who never read about databases. However, if you know what tables/indices are, get something else.

I'll preface my review by saying that my database skills are at least intermediate, if not advanced.

I started years ago by learning MS Access 97, and then some VBA. I moved on to SQL Server 7 and then 2000 for a couple years. I had not touched a database in at least a year when I was commissioned for a project that required more tables and relationships than I had previously done. That's when I bought this book.

I already knew concepts of normalization and how to write a SQL statement. This book gave me a fantastic review of good database design, without being specific to any one software approach. It taught me an accepted standard for drawing an entity relationship (ER) diagram and started me thinking about more complex relationships in a new way.

As a review, it was a quick read. I'm not sure it would help someone who's been doing databases for 10 years full-time. Nor do I think it is the absolute best beginner's text. However for someone at a beginning to intermediate level, this book may provide a much-needed review with several nuggets of new information.

"~I bought this book along with 3 other database titles because I needed some quick help in modeling a database. I am a relative newcomer to this field, but I've read a few other books and can pick up things pretty quickly. I was a little concerned about the price for 286-page book, but I found myself grateful for the lack of time-wasting filler. is on SQL; that is not quite accurate. The second half of the book, entitled "Practice" is mostly taken up by 3 detailed real world examples, starting with a description of the businesses to be databased, and ending with ER diagrams and SQL database definitions. The first 2 chapters of "Practice" review SQL and the capabilities of CASE tools, but most of it is taken up by the three examples. The examples are very helpful because they run into some of the same modeling problems I have"~ been stumped by, and they introduce things not covered in the Theory part of the book, such as 3-way composite entities and multiple relationship between the same 2 entities. diagrams with explanation. Harrington's book sort of splits the difference between these two other books, but all are helpful.