Customer Reviews

Expert One-on-One Visual Basic .NET Business Objects

5 star:		(17)
4 star:		(3)
3 star:		(1)
2 star:		(1)
1 star:		(0)

 

 

This book follows probably the most logical progression of any technically oriented book I've ever read.

Architecture and Design
Key Technologies
Implementing a Business Framework
OO Design
Business Object Design (using the Framework)
Windows, Web, and Web Services Interfaces
Reporting and Batch Processing

The premise of this book is that there are best practices that apply when building software systems. We've all heard that catchphrase before, but Rocky does a very good job of distilling it down to a practical level.

The book walks you through from proposed architecture to a fully functioning program, and along the way you learn some very powerful concepts:

Business rule tracking
Principal-based security
n-Level Undo
DataBinding
Remoting over HTTP
Reflection
Transactional methods using both COM+ (Enterprise Services) and native .NET OleDbTransaction and SqlTransaction
Lightweight collection objects
The true best use of web services

No-touch deployment

My favorite parts of the book were:

1. His approach to data access. Rather than creating a separate Portal object and forcing the UI to create two objects to access data, this framework places virtual methods in the base classes that must be overridden in the business classes. The system then uses Reflection to call back into the business object (from the Portal object) for the implementation of the data portal methods. The UI developer, however, sees none of this. Instead, the UI calls static factory methods to fetch business objects, and a very simple Save() method to add, update, and delete. This is a very intuitive approach, because the abstract nature of a Save() command is very comfortable to a UI developer.

2. Separate, lightweight objects for collections (for display in lists &c). Since this intelligent business object approach can create fairly "heavy" business objects, the framework has some great base classes for collections. Since you usually display only summary information in lists &c, why not create a specific lightweight object just for this purpose? Rocky shows you how to do this using structs rather than objects. This helps performance since they are a value type and stored on the stack. That may seem counterintuitive, but since this framework makes heavy use of serialization to pass objects across the wire by remoting, the gains from using reference types are mostly wiped out anyway.

3. His approach to web services is very practical. Rather than seeing them as a universal savior and placing them as interfaces between every nook and cranny of our code, he takes the approach that they are the "machine interface" to our code, rather than the human interface. This frees us up to develop the business functionality for a specific project, create the forms and/or web UI, then build a specific web services interface to that project/module when the need arises for an external API. He also talks at length about the foolishness of exposing our core business objects to the web services interface. If we do that, we lose the ability to change the interface, because the external contract has been established. Imagine, for example, if UPS suddenly decided to change its web services interface. Mass chaos. Rather, he shows how to create methods that are specific to the web services interface, and are more abstract in their implementation, thus less likely to need changes.

As a C# programmer, I was a little leery of buying this book, but I found out that by having to translate the code in the book, rather than just typing it in, I had to think more about the techniques involved. This helped so much, in fact, that I would now purposely buy books that are not in my language of choice, so that I can better understand the concepts instead of the syntax.

The amount of information provided by this book is enormous! Not only does Rocky explain how to build and use business objects in .NET, but he also provides a complete business object framework.

All .NET developers (not just VB.NET!!!) can benefit greatly from this book. Rocky provides insight and experience that would take years to accumulate, or hundreds of consulting hours to purchase.

Being new to .NET when I bought this book (a year ago), I was looking for architectural guidance. I wanted a layered architecture that fit well with object oriented programming. This book clearly hits the mark. The topic (Rocky's CSLA architecture) is great. It's based on some really sound principles, and has a noble goal: namely, smart business objects. And it gives good detail. While CSLA centers on the business layer, Rocky does a good job of explaining how it fits in with other layers.

Perhaps the thing I like best is the writing. Rocky is an excellent author. He stays on track, he's organized, and above all he is clear. He forsees many of your questions and answers them up front.

And Rocky puts his money where his mouth is. He addresses questions about the CSLA framework in the MSN email group on a daily basis. Let me see: a free architecture (he isn't selling a product, you're free to use the CSLA architecture in your work), an excellent book, and an author who's accessible and keeps working to advance the architecture in his free time!

If you have any lingering doubts, then go to a book store and pick up a copy and read the first two chapters. He does a fantastic job of explaining what he was trying to accomplish and how he went about it and the various design trade-offs that he encountered and how and why he chose to address each of them. If even skimming the first two chapters doesn't convince you it's worth the read, I'd be very surprised.

Finally, I would like to comment on an earlier review suggested that Rocky was retrofitting CSLA for VB.NET. I would like to respectfully differ. I think that CSLA fits BETTER with the .NET framework than it did in VB6. In VB6, he required several work arounds to problems that have been resolved in .NET, and these are clearly mentioned in the book.

p.s. The book was so highly sought after in the development community that he has written a C# version of it too, for those who are interested. It may be out already.

I have spent the last 12 months getting up to speed on .Net programming and read about 10 books. This one is by far the best.

What sets Rockford's book apart is that he solves problems by setting an objective, discusses approaches and highlights pros and cons, delves into the coding structure, walks through the solution in the chosen programming language, and gives adequate information/description on the steps in the IDE.

Other books I have read focuses on how Microsoft has implemented the technologies and libraries. Rockford puts it to use for a corporate programmer.

Many advanced programmers may disagree with his solution and point out weaknesses, but so does Rockford. The enormous benefit here is learning problemsolving with some advanced thinking and forethought, EVEN IF YOU DISAGREE with Rockford's answers and solutions!!!

The biggest value you'll get is learning to solve problems. BETTER YET, there are active discussions online at www.lhotka.net (Rockford's website) on solving real problems using his solution - even enhancing and adapting it.

Using .Net, the framework presented can be used in other languages for application development. Hence, you can learn to use the framework and progress in your professional work.

I'd say this is the book that brings you up a notch or two in the programming world. Real experience put to real use in the real world is the gyst of this book.

If you are not a VB programmer, get it anyway. If you are a C# programmer, get this VB version and buy the C# version to be released in May 2004!!

It's not easy to reach Rockford's level, but this book sure gets you going!

It will move you from intermediate to advanced in a month of dedicated work! A book for anyone who is serious about their professional career!

As you probably know from this review I give it 5 stars and top rating!

I'm an MCSD with about six years of development experience, mostly with web applications. I highly recommend this book to anyone interested in developing a very scalable business application framework. I'm not yet done reading the entire book, but I already can't wait to implement these techniques when converting my company's existing huge ASP application to ASP.net. Many thanks to the author!

I developed a number of apps. utilizing Rocky's architecture for VB6, so, when I was learning VB.NET and noticed Rocky had a book out on his CSLA architecture in .NET, I knew I had to buy it. As is the case with all of Rocky's books, they are the best value out there. He explains the stuff extremely well, (yes, I have to read a few of the chapters a couple of times, but, this is some difficult stuff) and follows up his explanations with solid, correctly executing code. I learn best by examples, and loading his projects and stepping through them provide me with an invaluable learning experience. I cannot wait for his C# book on CSLA. I have already been on his website and downloaded the beta code in C# and started working on some C# projects so when his book comes out, I will have already tackled the material.

Experienced developers know that there is a certain set of practices which aids in better n-tier development. For example, encapsulation of business rules, design for scalability and reuse and usage of design patterns. The first few chapters of this book do a good job of introducing these concepts in a way that makes them accesible to developers who are unfamiliar with them.

For the rest of the book, Lhotka describes the CSLA framework; a toolset he developed which is intended to hide many of the implementation details just discussed. While there are situations where his framework has it's uses, Lhtoka describes it as if it were the Swiss-Army knife of business development. Unfortunately, it isn't. What Lhotka has really discovered is a good way to build applications that works for *him*, but he seems determined that *all* developers would be better off if they all worked similarly.

Pro-golfers know that there is no such thing as a perfect swing. Every golfer is different, and the ultimate swing for one is completely different from the ultimate swing for another. Business development is the same. Each project and even each combination of developers will benefit more from coming up with a style that suits thier particular situation best rather than force-fitting a generic solution, such as CSLA, to thier individual effort.

Lhotka is an accomplished developer. Therefore this book would have been better positioned as a career-autobiography. In other words it's central message would have worked as: "I'm a successful person, and here's how I accomplished it". Instead, the message it delivers is "If this framework worked for me, it must work for everyone". Any pro-golfer (or experienced business developer) will tell you that this simply isn't true.

In case you're unaware of it's history, this book is meant to be Lhotka's update of his CSLA framework for .NET. Unfortunately, the fact that it wasn't designed for .NET handicaps it severely. For example, in several places, Lhotka uses "tricks", such as reflection to compensate for what he calls limitations of the .NET framework. In reality, what Lhotka is compensating for in these areas is simple poor design, especially the aforementioned fact that it doesn't really fit with the .NET framework design. This is unfortunate, because what it really shows is that CSLA would benefit more from a complete, ground-up redesign for .NET instead of an ill-conceived retrofit. A much better framework could be developed by embracing certain features of the .NET framework instead of trying to thwart them.

Inexperienced developers will probably pick up some good n-tiered component development practices from this book, but they will probably also pick up some bad ones, and unfortunately, being inexperienced, won't know which is which. Experienced developers will read this book, shrug, and move on to a more insteresting one. Someone from either group who is interested in learning more about n-tier business development would do better to pick up a book on pure theory, or dedicated to solving his or her specific business problem. Unfortunately, since this book tries to be both things, it ends being good at neither.

I don't know of any other author as dedicated as Rocky Lhotka in helping developers to most productively apply software engineering best practices to solve real-world problems. VB .NET Business Objects is the only text I have found that is successful in developing 100% of the instruction needed for a developer with a moderate to advanced understanding of OO development to become significantly more productive when developing enterprise solutions.

I award Rocky 6.0 out of 5 stars on his the .NET version of his text. (A previous edition covered VB 6.0 and an upcoming edition will address VB .NET 2.0.)

No single text can address all aspects of software development and this book far exceeds all or almost all others available. That being said (and without reducing the score I assign to Rocky's book), there are always improvements that can be made. Here are my recommendations for the next edition:

Readers who are new to OO development and who are using this text to understand best practices of applying OO techniques while they simultaneously learn OO basics from another source will find the text challenging. Rocky's book provides some basic OO instruction but is most beneficial to those with a fairly solid grasp of the underlying concepts. That statement is not intended to reflect negatively on the book, as no single book can address all needs.

One suggestion of how Rocky might expand the audience for his next edition, however, is to select a different case study for the book to follow. The current edition uses a Project Tracker case developed by Rocky for the book to illustrate the concepts he describes. Less experienced developers could gain more benefit from the book if Rocky selected a case study that was already described in a well-respected beginning-intermediate OO text. Rocky could focus on how to implement the case study using the CSLA .NET framework. Readers who are new to OO design could reference the detailed instruction in the corresponding text if they need a better understanding of the underlying OO concepts.

I am sure there are many case study options available for selection; I happen to like the POS case study used by Craig Larman in "Applying UML and Patterns: An Introduction to Object-Oriented Analysis and Design and Iterative Development, Third Edition."

With such an approach, Rocky could produce a volume that would be as beneficial to beginning developers as his current text is to intermediate and advanced developers. The level of support that Rocky (along with the large and responsive VB .NET BO community of readers) graciously provides would likely be significantly reduced if an existing text answered many readers' basic OO questions.

This is by far one of the best .net books ever written. Lhotka manages to bring together all the pieces of .net and show how they can be put together and deployed in the real-world. I truly believe that this book should be issued with visual studio. If you want to build scalable multi-tier apps, buy this book. You won't regret it.

P.S. Get this book while you can! Its ashame that the publishing company Wrox went under and its titles were liquidated to several smaller companies. Currently, Wiley has the wrox name, but they only bought rights to like 30 of the 300 titles.

Rocky's book is a MUST HAVE!
His no-nonsense approach is straight forward and down to business. No fluff here, just PLENTY of meat.
His concepts are clear, precise and right on target.
Even if you are not currently design the next generation killer ENTERPRISE APPLICATION, his advise and knowledge are invaluable.

key words: INVALUABLE & MUST HAVE !