One of the most valuable concepts in the book is an IT organization that is defined by technology layers as opposed to products. For example, a compelling arguement is made for organizing the systems administration function as a single group without regard to what brand of system is being administered. The same argurment applies to organizing DBAs, network administrators, etc. in the same manner.
This is a powerful concept that has a lot going for it. For example, in the traditional organization system administration is performed by a number of groups, each focusing on NT, UNIX, etc. This promotes a disjointed and non-repeatable set of processes - if there are processes at all. This, in turn, leads to an IT organization that has no clear internal communications, a cacophony of wildly different processes and methods, and multiple agendas. It reinforces the business side's common complaint that IT of out-of-control, with no unified vision, as well as another often heard complaint that IT provides conflicting advice and are their own worst enemy.
Contrast the above with the organizational model that is proposed in this book: all functions are grouped and held together by a common set of processes and procedures. One easy-to-spot advantage of this type of organization is that service delivery becomes easier. Problems such as synchronizing batch processing (essential to data warehousing), aligned maintenance windows and uniform approaches to problem management become manageable because everyone is on the same team.
Another advantage is a leveling of process maturity. Mainframe administration processes are lot more mature than those employed by your typical NT administrator, who would benefit greatly by "discovering" what was probably in place before he or she was born. And the business - the real reason we IT professionals exist at all - will benefit from the improved and reliable delivery of services and support.
There are gaps in some of the processes and organizational paradigms, as pointed out by other reviewers. These will require some thought on the reader's part to work through and fill. On the whole, however, I found the book to be a valuable source of concepts and ideas. The flaws and gaps are offset by some iteas that I though were excellent. Because I personally gained a much deeper understanding of how to align IT to better meet business needs I gave the book 4 stars (only because I cannot award it 3.5). In spite of the flaws and gaps I do highly recommend this book and hope that potential readers will look beyond the warts and find the enlightening information buried between the covers.
ABOUT ITS VALUE: The author has succeed reorganizing IT departments and he wants to write about the importance of applying mainframe administration paradigm to client/server solutions. The book is helpful only if the reader wants to know what could be wrong at the IT Department. Although there is a proposed model, there aren't specific recommendations, choices, roadmaps, deployment guidelines, impact analysis (budget, time, employee morale, issues, risks, etc.). The model isn't complete. The author doesn't explain how to structure and integrate applications development teams, corporate applications administration, decentralized IT support personnel, and outsourced areas among other important functions. The project management function is mentioned but its explanation is avoided.
If you really want to build a world class infrastructure look to _IT Systems Management_ by Rich Schiesser. It's also in this series and is everything this book is not.
_IT Systems Management_ does not really cover desktop support/helpdesk issues, its one minor shortcoming. For that look to _IT Problem Management_ by Gary Walker, also in this series.
You'll find both _IT Systems Management_ and _IT Problem Management_ here at Amazon, and they are both highly reviewed and they will be much more helpful than this book.