Common Warehouse Metamodel: An Introduction to the Standard for Data Warehouse Integration
||Author: John Poole, Dan Chang, Douglas Tolbert, David Mellor|
List Price: $34.99
Our Price: Click to see the latest and low price
Publisher: John Wiley & Sons (15 November, 2001)
Sales Rank: 77,148
Average Customer Rating: 3.86 out of 5
Customer ReviewsRating: 2 out of 5
Worryingly vague and unimplementable
I'll say at the start that this is not my kind of book. I prefer books which are useful, enlightening or both. This didn't seem to be either. From page 3: "The mission of this book is to provide a single, coherent, and comprehensive overview of the OMG's Common Warehouse Metamodel, which is easy to read.". It may be slightly easier to read than the raw specification, but it's a lot less useful. The most telling point is further down the same page where it admits to really being just an introduction to a forthcoming "Warehouse Metamodel Developers Guide".
For an overview, the book is really short on examples. It's got lots of vague UML diagrams and pretty pictures like you might see on a powerpoint slide, but not a single worked example to show how all the buzzwords and technologies might actually fit together. I also have great problems with their use of UML as a language to actually specify data models, processes and so on. For me UML is a tool to help express intentions to people, not supply details to processing software, but this book seems to ignore the difference.
If you know nothing about meta-modelling, and want the sort of information you can get from the slides of a conference presentation, this may be a useful book. If you want to understand the details, or (gosh) actually get a job done, then this book will just frustrate you.
Rating: 5 out of 5
Good overview for good technology
I can't believe the previous reviewer. He basically said he didn't like CWM or UML, therefore the book is obviously bad ?!?
The project I am just starting is a large data mining effort that will be integrating multiple data warehouse and data mining tools. I knew we needed CWM from some earlier work with metadata repositories, but did not have the energy to dig into the OMG specification. This book gave me exactly the overview I was looking for; as an earlier author said, "This book covers all the practical steps for planning, implementing, and deploying CWM technologies". I would like to give it at least 7 stars to average out the previous irrelevant review...
Rating: 1 out of 5
The emperor has no clothes
Ok, I am a known heretic. I am not impressed by the CWM model. It is oriented toward the object-oriented implementation of a tool for metadata exchange, not toward representing the things business people would be looking for in a meta data repository.
This book is better than the on-line specification at describing the model--which was really incomprehensible--but this is at the expense of completeness. Definitions are not available for all classes and the ones that are are not clear (to me at least). The relationships are barely defined at all.
In fairness, the model is so complex that it may not be possible to describe clearly to anyone not deeply immersed in the language of object-orientation. The team of authors is further hampered by its use of UML. The notation does not permit a complete inheritance tree to be portrayed in a diagram if the diagram is of less than the entire model. Two classes may be related, but you can't see this because the relationship is between great grandparents, shown on a different page.
· Model Driven Architecture: Applying MDA to Enterprise Computing
· Metadata Solutions: Using Metamodels, Repositories, XML, and Enterprise Portals to Generate Information on Demand
· Building and Managing the Meta Data Repository: A Full Lifecycle Guide
· Building the Data Warehouse (3rd Edition)
· Common Warehouse Metamodel Developer's Guide