on May 28, 2008
This book is great stuff for Enterprise Architects. The discussion of Zachman is better than any of the Zachman Institues articles. The explanation of the value of architectural meta-models is excellent.
On the down side, the meta-models presented are very good EXCEPT that the author still presents data as being a part of an application. Surely 20 or so years after James Martin we are past that. Applications with their own data schemata are to be avoided and suppressed, rather than endorsed.
If your are a "real" Enterprise Architect, then this is book indispensable, but review all of the meta-models carefully to insure that they comply with your particular religion.
on November 17, 2006
David Hay has produced another great book for the Data Architect, Data Administrator, and others who seek to understand data models, patterns in models, and metadata. It's more abstract than the previous Data Model Patterns, Conventions of Thought, and the focus is on information about information or metadata. The book is based on an Architecture Framework that is derived from John Zachman's "Framework for Enterprise Architecture". Examples are well developed and understandable. Much of the book expands on work and discussions through the Business Rules Group so you know the content is rock solid.
I especially liked the Entity Relationship Diagram (ERD) for the EntityType, Attribute, and RelationshipEnd. This simple but comprehensive model allows for super and sub types and recursion. The Class Model (UML) with its components parallels the similarities for the object oriented perspective. One of the fundamental strengths of this book is the ability to view metadata from different perspectives. "Gemba" is a wonderful word from the worlds of Total Quality Management, Lean Thinking, and Six Sigma. It means "real place". This book is "gemba" for metadata. The different views or perspectives of metadata are excellent and perhaps that's why it fits the Architecture Framework so nicely.
Within each chapter, David Hay discusses a column of the Architecture Framework. The chapter on people and organizations is awesome! He covers the business owner's view, the architect's view, designer's view and security and governance. Significant numbers of easy to follow color-coded diagrams accompany each section in all the chapters that clarify and elaborate on the text discussion. At each level, the topic is put into the Zachman framework for added clarity.
The chapter on the motivation column is also exceptional. I've never seen ERDs for mission, vision, objective, goal, desired result, means, tactic, strategy, directive, fact, business rule, business policy, enforcement, and consequence. How outstanding! That chapter alone is worth the price of the book and it really got me to thinking in much larger metadata type terms. Those organizations faced with governmental and business policies such as Sarbanes-Oxley will have much to think about with this data model!
Back in 2002, it was my distinct pleasure to read for the first time David Hay's book, Requirements Analysis: From Business Views to Architecture. That book is excellent and is packed full of features that really impressed me. He's included those same terrific features in this book as well. So as I wrote then, I'll repeat now: "The index is excellent! It's quite extensive, complete, and well organized. The bibliography is also exceptional and very complete. The glossary of terms is first rate! I especially liked the notations of chapters where the terms were documented. All of these excellent features point to the superb organization of the book. It's a classic read and a must-have reference!"
Maggie Tompkins is a lead designer/developer for corporate database projects. She's a member of the Oracle Development Tools User Group (ODTUG) Board of Directors and is Editor Emeritus of the ODTUG Technical Journal. She is a previous vice president and secretary of ODTUG. She won the Best Speaker award for the ODTUG 2000 conference. Maggie has thirteen years of experience with Oracle tools and expertise includes Oracle SCM, Designer, database design, and methodology. She's a teacher, mentor, frequent presenter, and active participant on the ODTUG list serves. Margaret.Tompkins@comcast.net.
on April 1, 2010
This is a very good book with a somewhat obscure title. Its subject is how to construct an information management system for a commercial enterprise. Its sub-title, "A Metadata Map" actually refers to the fact that the first step in building such a management system is to describe the business of the enterprise for which it is being constructed.
What Hay has done is update the absolutely first rate work of John Zachman who over twenty years ago developed the "Enterprise Architectural Framework" which is essentially a matrix that can guide the development of an information management system based on the organization structure and processes of the enterprise itself. The Zachman framework is the basis for the "Data Model Patterns" in the title of this book and indeed the book is organized around this framework. It provides step by step descriptions on the construction of an information management system for a notional enterprise.
It should be noted that Hay is well aware that information management is more than simply building a database (data warehouse). Information storage is pointless without some means of organizing and classifying it. In his experience as a consultant to Oracle he learned to model the structure of an enterprise not simply a data base. This is something that is possible using a relational database which can be used to describe the structure of an organization. The "metadata" used in this book refers to a description of the information used and to the description of the enterprise itself. As Hay notes there are many resources that can be used when constructing an information management system for a commercial enterprise. For example he advises going to the Business Rules Group to obtain both correct terminology and rules compliant with standard business practices.
This book is good as a practical guide and as an excellent exposition on the Zachman Framework.
on January 11, 2007
I have always considered Data Model Patterns by David Hay to be a classic. In this new book, he extends his perspective to cover the questions raised by metadata, providing answers in the context of the Zachman framework.
Hay show how far simple reasoning, clear language, and outstanding diagrams can go to help document enterprise architectures. There is no need for fancy mathematics, normalization theory and predicate calculus to capture the essence of an enterprise.