I feel the book is weak in terms of focus on the exam itself and the sample tests, etc. Many of the exam questions were very poorly worded. They don't give you enough information to make a really good decision. In some cases they were just plain wrong. (Fortunatly, I didn't find that to be the case with the real exam.) One really dissapointing thing is the way the big practice test at the end nearly repeats a lot of the individual chapter test questions.

So, the information in the book is good but somebody who really knows testing methods needs to rework it.

Also, I think the book emphasizes areas (e.g. replication, full-text search) that aren't stressed very much on the exam. Conversely, it seemed not to cover other areas (security and performance monitoring) very adequately.

This book needs to be proofread thoroughly and corrected.

The CD only contains a softcopy of the book itself plus some kind of sample test. (Since the sample test in the book were so poor, I didn't waste any time installing and taking the test from the CD.) A very useful addition would be a CD containing enough data to build a reasonably large database. This could be used to give practice in creating, loading, and then running performance tests, etc. "Northwind" and "Pubs" just aren't big enough to do this.