12 of 12 people found the following review helpful
on 16 June 2006
Problem with this book is it is a bit woolly and wordy (just like the previous reviewer described). However, the main difficulty is there simply is no other book around on the market. As a description of the entire end-to-end ETL process, including many subject areas I'd not even considered (eg. COBOL copy books), it's very good.
However, I'd say the REAL reason for buying this book is it works well with Ralph Kimballs other work "The Data Warehouse Toolkit", and gives an excellent summary of Dimensional Design. I guess the authors felt they must put this in to explain the background. Personally I found it invaluable.
Also the description of "real time ETL" was invaluable. Everyone's talking about it, but the book gives a credible outline solution.
Yes woolly, yes it uses 10 words when two would do, but overall I got a lot out of it.
32 of 33 people found the following review helpful
Computing is an exact and unambiguous discipline; consequently I want my computer books to be written in an exact and unambiguous manner. "The Data Warehouse ETL Toolkit" falls far short of this requirement, being wordy, vague, overblown and crammed with jargon. Worst of all, I found there were very few "Practical Techniques" I could take away with me that would help me in my work.
Here's a sample sentence: "This section discusses what needs to go into the data-cleansing baseline for the data warehouse, including simple methods for detecting, capturing and addressing common data-quality issues and procedures for providing the organisation with improved visibility into data-lineage and data-quality improvements over time". Now imagine a whole book written like this. OK, I've taken this sentence out of context, but if I tell you that this was used to introduce a section - there are no preceding or trailing sentences - then I think I am starting to paint a picture.
The authors and publishers seem to have taken the attitude, "Why use a bullet point when a paragraph will do?". Text and examples have been embellished as if in an effort to prove how clever the authors are. A lot of jargon is employed (no glossary), but the reader is always left in doubt as to whether this is industry standard or idiom employed only by the authors.
I think this book could have been so much more useful if they had taken a worked example right through from start to finish. They could have explained where the real world may be different to this perfect model and drawn on their experiences to add colour. Also, if this truly was supposed to be a book of practical techniques, they should have highlighted them, say 1 to 100, through the text, as applicable.
So why two stars rather than none? Firstly, because there are some good nuggets of information in there, if you work hard to find them, and secondly, because the authors' interest for the subject does show through. They do have a knack of answering the important questions, but only after a long journey round the houses first.
Kimball and Caserta would probably be fantastic consultants to have on a big data warehousing project, unfortunately they are awful technical writers - only buy this book if nothing else covers the subject you are interested in.