Customer Reviews

Expert One-on-One J2EE Design and Development (Programmer to Programmer)

5 star:		(29)
4 star:		(5)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

First of all, this is a fine book. It is loaded with valuable nuggets and insight that I have yet to find in similar books. I agree with all of the other reviewers that Rod's style is great and that his strong opinions (based on obviously extensive experience) are a welcome relief to the simple re-organization of Sun's specification and/or design pattern recommendations.

ONE CAUTION! Some of the framework code outlined in the book COULD be considered overly complex for many circumstances. Ironically, Rod would probably respond that the high degree of abstraction would IMPROVE the usability and maintainability of any J2EE application. However, I believe that excessive abstraction, in an of itself, can become a maintenance (and training) nightmare (look at the sheer complexity of the interface inheritance in the data access framework from Chapter 9 alone). His solutions to Data Access and Application infrastructure are so sophisticated that they qualify as a framework worthy of open source momentum on their own. The downside of this, of course, is that your development team must invest in understanding these sophisticated hierarchies since they are unlikely to have encountered them in any previous role. Unless you have a VERY stable team of highly skilled designers/developers, be cautious of trying to implement such frameworks. Instead, the downsides of more simple (albeit less flexible) approaches are usually outweighed by the reduction in training or the likelihood that transient resources will be able to contribute more quickly. Nonetheless, a great book and worthy of my ** All-Star ** category.

Even though this book is named "J2EE design and development", it seems to be much more than that. This seems to be the book in which the author attempted to put, in a crystallized form, most of his expertise not just in J2EE (that would be trivial), but in programming and (particularly) architecture in general. Given that the author is a true expert in the field (this is clear after reading just a few pages), this book has a value beyond anything I can express here in my words. I learned from this book more than from any other book on programming and architecture, with a possible exception of GOF "Design Patterns" classics.

No other book on web programming that I know of comes even close to this one. Some noteworthy features:
-- Always framework-oriented approach (which in my view is the only possible choice for real-world projects)
-- Heavy emphasis on architectural side of web development (follows from the previous point)
-- Comparison of different view technologies from practical point of view, w/o exclusively subscribing to a particular one which seems popular (like, JSP).
-- Excellent coverage of MVC paradigm, again, w/o subscribing to a particlar implementation (like Struts, etc)
-- Extensive coverage of all levels ("tiers") of a web application.

But what really shines, are the insights on architecture, namely the things which is impossible to find out by theoretically studying J2EE specifications and books such as "Enterprise Java Beans" and the like. For example, why Entity EJBs don't work. Or when is collocated EJB architecture is more appropriate than the distributed one, and why. After reading the book, many concepts just clear up. Not to say that everything is written in a clear and concise language (despite a few typographical errors that Wrox books are notorious for).

In short, this book is simply amazing.

The real differentiator of this book is that its focus is on how to use J2EE to solve real-world problems, rather than providing an overview of the J2EE specification itself. As such, it's a much more pragmatic approach to using J2EE than the other books I've read on the subject.

The author starts by taking high-level perspective on the subject, showing the different design considerations that need to be applied when choosing which J2EE technologies to use. A case study (a ticket booking system) is described and elaborated on throughout the book, showing how these design considerations affect a real-world solution.

The book also provides a load of coding tips I found useful, from better use of reflection, to judicious use of design patterns, to how to minimise your refactoring. The author assumes the reader is an experienced developer, so doesn't focus on rudimentary Java and architecture. I found this useful, but it means the book is more focussed to a specific audience.

Amazingly, the book also provides an entire framework based on J2EE on which you can put your application logic. My only complaint is that this is not included in a companion CD - you have to download it.

Overall, the book works because it shows you how build solutions, not just understand the technology.

This is a great book - by far the best J2EE design book I have read to date. Approximately the first half of the book contains not just the HOW, but the WHY of J2EE design - this is exactly what I was after, and this book certainly provides it.

My only (somewhat minor) dissapoinment with the book was the fact that a lot of the examples were not built around existing frameworks, but were based around Rod's own frameworks. For example, instead of using Struts, Rod uses his own MVC web framework. Even though Rod's frameworks are probably *better* than some that are out there - it would have been nice to read "portable" examples that can be plugged into existing frameworks.

Having said that, I still think the book is an excellent buy, and a must for any serious J2EE developer/designer/architect that wants to understand WHY you should architect your J2EE system in a particular way - not just HOW.

I never taken the time to write a book review on Amazon.Com before, but after purchasing this book, I thought I should. This is the most practical book I've read yet on developing J2EE applications. As a J2EE application architect, I must say it addresses all the issues I've encountered over the past several years. I wish this book would have existed a couple of years ago (so I wouldn't have had to learn the hard way). The book really focuses on doing what's practical and discourages over engineering due to J2EE hype. The book is very current and discusses many of the widely used open-source products -- something I'm a big proponent of. In my mind, this book should be considered the "J2EE Bible". Lastly, you don't have to be an application architect to understand this book. It's very easy to read -- obviously intended for J2EE developers at all levels. I'd give this book 6 stars if I could.

This is the most useful book I have read so far on best practices in developing web applications and J2EE enterprise applications.

The problems we try to solve with web applications are varied and complex, and J2EE technology generally makes creating a solution very difficult for the unwary. J2EE provides a lot of complex, suboptimal alternatives for the parts you need to put a web application together, and this book provides much needed guidance in how to get to best practices and how to avoid blindly following bad advice from the pundits. The chapter on how basic OO practices relate to J2EE applications is worth the book, and the advice on persistence, use of JSP and other web tier technologies, use (and non-use) of EJB, and classloading are all excellent.

But it is not a perfect book; here are a couple of caveats. J2EE technologies and best practices move fast, and almost all of the examples in this book use Rod's earlier framework that grew into the Spring framework, so the examples are amost all dated. Also dated is Rod's coverage of JSP, as this book was written before the release of JSP 2.0, and Rod does not cover the shift in best practices for JSP that the newer release represents. He does hint at new practices however, and points out the potential of the (at the time) upcoming JSP 2.0 release.

One more note, this book covers different topics than the more recent "Expert J2EE Design and Development (Without EJB)," and that book (which covers Spring among other topics) refers back to this one a number of times, so even though this book is dated, it is still quite useful in addition to, not instead of, it's companion volume. (..And that's my own opinion. I'm not affiliated with the author or the publisher.)

This is the one if not the only book on the market that can truly serve you as a comprehensive manual for J2EE solution architectures. Every line in this book is worth of gold. It personally helped me justify, reinforce, discover or solidify some very important architectural desicions in my practice.
For instance there is a whole section on presentation tier technology choices. That section covers all of the popular frameworks and technologies (JSP, Struts, XSLT,...).
Each technology is described in terms of what it is, and what are its benefits and drawbacks. Then there is a very good code samples section. Author uses one application throughout the book, and then implements it using various technologies.
Moreover, he suggests you when does it make sense, and when does it not to implement the technology as a solution. It is amazing how much wisdom is built into this book.

Of course some of the APIs covered in the book will be outdated (EJB 2.1), but that does not bother me much. The wisdom is what matters.

Writing in general is very thorough, very practical and reinforced with some very strong real life examples.
Author obviously posseses the maturity and experience that
is so rare to find.
It is a great professional resource, and career builder.

This is exactly the book I needed to design a wJ2EE based web delivery system - Rod Johnson has done the developer community an *exceptional* service by sharing/communicating his experience and insights in J2EE development so effectively. The "working through" of the various design alternatives and the decision process in a thoughtful and balanced manner is essential to the success of the book - in particular I found it most useful to consider the pros and cons (in some detail) of the various options in the "universe of possibilities" before making a final decision on specific implementation technologies - Rod Johnson has distilled his experience in the field so that senior developers such as myself may build on his experience rather than repeating it. My sincere complements to him for a job well done.

Rod Johnson is one of the few technical authors with whom I can almost never disagree. A quick read indicates clearly that his technical insight, which ranges from architectural to low-level coding best practices, are not born of some academic exercise...they are the fruit of actual production J2EE experience...not an academic blueprint. At times, I felt like I was reading my own words. Over the years, I began to wonder if I was the only J2EE developer who was not "drinking all the kool aid." My experience with over a dozen high-volume production applications moved me away from the pure party line. Now, I realize that my religion has a leader. Don't get me wrong, I learned a significant amount from this book. Rod's experience is daunting and even an experienced J2EE developer will glean countless insights from this well-written text.

So what's not to like? Well, frankly, I was disappointed that security got the same level as attention in this book as it does in most - especially since there has yet to be an excellent J2EE text produced on the topic. While I didn't expect Rod to write the definitive tome on authentication and authorization, I expected more than two pages with a collection of URLs for more info. In fact, I loved the fact that he led off the text with testing and was shocked that he didn't follow immediately with security - another system aspect that is frequently relegated to the margins...and often implemented poorly. So how does that influence my review? Well, on Amazon's five star scale, I am taking away one star....but I also started by awarding him ten stars for the rest of the text.

final static int MAX_RATING = 5;

final int rating = Math.min(MAX_RATING, (10-1));

if (rating == 5) {

you.buyNow();

}

Rock on Rod. Can't wait for the "Developing without EJBs" text.

I appreciated the clean and clear style of this book. Important points are highlighted within the shaded boxes of text found throughout the book. It is apparent that Mr. Johnson has spent many hours working through problems with the existing J2EE architectures and he explains what he feels is worthwhile and what is not. I appreciate that this is accomplished without a lot of elaboration.

Mr. Johnson presents a practical approach to J2EE design with an in depth analysis of the Web-Tier Model-View-Controller design. I found the reference information helpful to assist me in preparing to grasp the dense subject matter presented. A review of MVC from the GOF was helpful as well as a review of Core J2EE Patterns with attention to the Service to Worker and DAO patterns.

Since Mr. Johnson has worked in the development of a MVC famework, the framework he has written is presented. The com.interface21 framework is presented with its infrastructure as he walks through the various design alternatives all the while guiding the reader away from design pit-falls while maintaining a clean delineation of responsibilities within the MVC framework.

Some of the diagrams are just too small to use without a magnifying glass but the diagrams are crisp with magnification.

This book is a must for the framework designer, or pragmatic developer that must implement J2EE architecture.

-Ralph Burroughs
January 12, 2003