|
6 of 7 people found the following review helpful:
5.0 out of 5 stars
It didn't stop there!, 28 Jun 2005
This book covers essential concepts that are even more essential to understand than they were when the book was written (circa 1977). In the intervening years, the consequences of poor data management have become much worse. The good news is that help is on the way! In the last 25 years, solutions have been developed to many of the data problems described by Kent. Sadly, it seems that very few developers or analysts really understand the issues raised by Kent which is probably why so many software projects end up as expensive failures. Chapters 1-9 expose the data management problems and the awful limitations of what in 1977 passed as "data models" (e.g. the necessity to force fit "many to many" concepts into the inflexible "1 to many" structure of the traditional heirarchical databases and the aparrent lack of any widespread understanding of conceptual models and abstract domains. (Still true today) Chapter 10 gives interesting insights into the relational model which at the time (1977) was only 8 years old and was not widely available in database products. Chapter 11 is entitled "Elementary Concepts: Another Model?" In this excellent chapter, Kent explains n-ary relationships and shows how binary relationships are best seen as instances of the set of n-ary relationships. Chapter 12 contains insights into the hidden effects of language on thinking. Many of the ideas in Chapter 11 later appeared in NIAM (Nijssen's Information Analysis Methodology). In the mid 1980's Terry Halpin worked with Sjir Nijssen and formalised NIAM as ORM (Object-Role Modeling language). Terry's most recent books on this matter are: 1: "Information Modeling and Relational Databases" -MKP 2001. This 754 page book is the ORM Bible and is a set book for University Students in the USA. 2: "Database Modeling with Microsoft Visio for Enterprise Architects" MKP 2003. This book shows you how to use the ORM tool which provides solutions to most of the problems that Kent describes in his book. This book is also a set book for University students in the USA. (I'm pleased to have co-authored this book with Terry) Historical Note: An ORM tool was developed in 1989 and first appeared as a product called "InfoDesigner". This tool evolved through the 1990's as "InfoModeler" which was bought by Visio who incorporated it into the high end Visio product. Microsoft then bought Visio and the fruits of Terry's many years of hard work and dedication are now (fairly deeply) embedded in Microsoft Visual Studio.NET for Enterprise Architects. (You can download the latest beta from the Microsoft Website). In conclusion: "Data and Reality" gives an excellent description of the horrendous data definition problems that are still with us today. If you read "Data and Reality" and then say "Wow! Yes! William Kent has articulated problems to which we urgently need a solution!, then I strongly recommend that you investigate ORM.
|