Customer Reviews

Service-Oriented Architecture: A Field Guide to Integrating XML and Web Services (The Prentice Hall Service-Oriented Computing Series from Thomas Erl)

5 star:		(24)
4 star:		(5)
3 star:		(1)
2 star:		(1)
1 star:		(1)

 

 

I hesitate to say anything negative about this book, given the glowing reviews by most other readers. Many of those reviewers appear to be experienced in SOA deployment. I am fairly new to the field, though I have a decent grasp of XML and Web services. I wanted a good guide on how SOA was an advance over a loose collection of Web services.

This book is chock full of information on the standards for Web services and the ingredients for SOA. Part one, on XML and second-generation WS specs, is excellent.

However, potential readers should know that this is an overview of these specs, not a tutorial or implementation guide. Certainly too many of these specs exist to give details of implementation, but if you're looking for a general guide to the specs, this is a good text for that.

Things get more muddled after that. I hoped that the chapters in Part II on integrating Web services into applications would be a practical guide to this task. I found it difficult to relate the abstract discussion and diagrams to how one would actually perform this difficult chore.

Later on, the book suddenly introduces EAI as if the reader would naturally know this technology. It uses EAI and similar concepts in more rather abstract discussion of principles of SOA. The discussion features few real-world examples about how the principles would translate into action.

The chapters toward the end on best practices are quite good and worth reading. Again, though, the discussion avoids getting into examples and details.

I kept having the feeling that the bottom line was that, to properly implement SOA, you need a consultant who knows the field well...and that the author's firm might be happy to provide that consulting. The SOA adoption methodology described in the book was basically what that firm follows, I gathered.

I suppose in a sense, the "field guide" name is appropriate here. A field guide to birds, for example, typically lets you identify the species so that you can say, "I saw this bird". It doesn't tell you the life history of the species, how to conserve it, or its ecological relationships to other species. This book has a similar role: it enables you to recogize SOA acronyms and understand the basic process for how an SOA operates and might be implemented. It will not give you enough to actually go out and build an SOA or fully evaluate SOA-related software. Perhaps the other book in the series provides more examples and completes the picture, though to get know-how on actual implementation I suspect you'd need more material.

If you think about the software development world as architects and engineers. Where architects take a very high level view of the world and don't get into fine grained implementation details. Then if you consider yourself this kind of architect, you will get a lot out of this book.

Though the book is fairly long (~500 pages) the depth of the content is still at the 'field guide' level. This means that the book focuses more on understanding the components of SOA at a holistic level without getting too deep into implementation details.

The first chapter of the book does delve into the basics of the XML core technologies (XML, XML validation, XSL, etc.). After that the book stays at the high level, describing most of the concepts with graphics that do an excellent job showing the document flow between systems.

I recommend this book to architects involved with XML based systems integration projects. I also recommend the book for engineers involved with these types of projects because they will benefit from the high level overview of the entire range of XML technologies.

I had high expectations when I ordered this book, and just having completed it, I felt like taking the time to share my opinions. I've read numerous computer books on various subjects. Most follow predictable form, and there's nothing wrong with that. These types of books meet an educational requirement and the more predictable they are the easier it is for readers to use them. This, however, is not such a book, and nor does it claim to be. This book calls itself a Field Guide - a guide you would want with you when you are in "the field". As such, it is structured with quick reference in mind. Not the type of quick reference guide you'd use to look up language syntax or reserved characters. This is a reference guide that you look to for cold, hard advice. If you are struggling with the many new issues that SOA and XML are hitting all of us with, you would use this guide to look for answers. And, on that basis, it really delivers. The author clearly has a depth of expertise that would cost an arm and a leg right now to hire. He has spilled his guts with this book, sharing not only product knowledge but also insights. The real benefit of this book is the insight, because that represents a body of knowledge gained from experience. That type of knowledge is hard to come by, which is why I can see this guide becoming valuable to organizations who lack experience with the whole XML/WS/SO field. I'm reluctant to give any book five stars. I feel that rating should be reserved for books that achieve unparalleled levels of excellence. I gave it four stars, because I feel this book is very, very good. It sets out to cover a cross section of organizational and technical areas that are most likely to be impacted by the arrival of SOA and all that comes with it. It blends strategic advice with best practices and an abstract exploration of common architecture blueprints. What the author has chosen to cover is appropriate and the manner in which he communicates the subject matter is efficient. I have no quarrels with recommending this guide, and I know I will continue to reach for it as new issues come my way in the future. If I had to change one thing about this book, it would be the location of the SOA modeling tutorials. For some reason they were placed at the end of the book, away from the other tutorials. I think a knowledge of SOA fundamentals up-front would help readers better understand the rest of the guide. Finally, I'd like to comment on what this book is not. This book does not talk about specific programming languages or middleware products. It sticks to standards, common architecture and general best practices. I find that approach appropriate for the world of SOA. SOA is fundamentally about neutral standards and platform independence. That makes this book also useful for just about any environment. Regardless of what vendor platform you are currently subscribing to, most of the information expressed in this book will be relevant, or, at the very least, of interest. This is equally useful from a learning perspective. Learning about XML/WS/SOA, middleware and integration without having to learn about the specific characteristics or unique features of commercial products gives you a reference point and plenty of ammunition for when you actually need to assess the product marketplace. Obviously, if you are working in a Java environment, you will want books on Java to build your systems. But when you design your system, I'd reference this book first. It helps you design a better architecture in abstract, before you implement your system with whatever development tools you choose. In other words (and to finally conclude this review), this book will not help you build Web Services. It will help you prepare for them, design them, position them, and integrate them.

This can be considered the defacto reference for Service oriented infrastructure setup initiatives and approaches.

Thomas Erl has made it a masterpiece with lot of positives, negatives and reasons for different choices that can be considered. First couple of chapters dwell into first and generation of web services including BPEL4WS, WS-S, WS-coordination etc. There is also explanations of strategic approaches of XML and database integration.

In the middle of the book, there are details about SOA and legacy integration and SOA and enterprise integration. Later parts of the book gets into best practises for integrating XML and integrating web services into the overall enterprise stack. All the SOA entities are shown in vivid details pictorially.

This is one of those books written with intent to help the readers with all the possible perspectives(both positive and negative) of the SOA. Great piece of work.

Very good reference for designing web services. Takes many low-level considerations into account, relating mostly to XML data formatted documents. What I also liked about this book is being able to read about web services without references to specific programming languages. It gives you a good grasp of concepts that you can take into development with you. This book should appeal to anyone wanting to get a broader perspective of the web services platform. It also has the best descriptions of service oriented integration architectures I've seen. Erl's loose writing style makes some of the more complex subjects easy to get. Thumbs up from me.

We are struggling with webservices/soa migration issues. Two of our systems are causing us a lot of problems. This book helped us with planning and education. It gave us different ideas for approaching migration and integration. This book gave us a migration strategy that is working. Not bad for forty dollars!

I would recommend this book to anyone interested in learning how webservices can be positioned to enable integration. The author provides many sample architectures and scenarios. A lot of detailed XML issues are also covered. These are more suitable for developers, because they deal mostly with optimization and modeling techniques. Also useful to us was the step by step process for designing soa friendly components. I've never seen this anywhere else, and I would imagine it is extremely important right now when so many of us are still building distributed apps w/o webservices. This book is pretty light on the tutorial end. But it doesn't claim to be a tutorial because it focuses on real world issues. If you are new to all of this I'd recommend you start with a webservices tutorial first. The author is also releasing a tutorial-type book on soa next year which I will buy. Also, this book has a decent support Web site (www.serviceoriented.ws) with a glossary and a diagram symbol legend that is not in the book.

Truly remarkable collection of tips and tricks and best practices covering the currently evolving SOA landscape. This guide focuses on the many ways and the many levels XML and Web Services can be integrated within the context of SOA. As such, this is not a book about theory as much as it is about dealing with real world problems.

I am mystified as to why people are comparing this book to tutorials or books about Java or .NET when this book clearly states that it is an integration field guide for SOA, XML and Web Services only. As stated on the back cover, this book focuses on the common, vendor neutral ground that is established by SOA. If you are looking for a book about Java, then get a book that has "Java" in the title! If you are looking for a tutorial or a book with case studies, then don't buy a book that brands itself as a "Field Guide to Integration"! I think some of the confusion has to do with people's lack of understanding about SOA. If you don't get what this book is trying to accomplish, then you don't understand SOA.

I am a Technical Architect for a progressive IT company. We have been working with Web Services for two years now, and this guide has helped me a great deal. I feel as though the author has accurately gauged the state of the IT industry's move to SOA. Although there is obviously an interest out there in SOA, most developers and architects must still contend with XML and Web Services within regular distributed architectures. For me anyway, the emphasis placed on XML, Web Services and SOA individually in this book is just right.

There has been a tremendous buildout in Web Services standards over the last two years. Its description forms the bulk of this book. The book factors neatly into two parts, which are interspersed as different chapters. One part is the above. The other deals with XML.

Consider the XML part first, because that is simpler, and Web Services build upon it. Even if you have no intention of using Web Services, this book has excellent advice on XML development. It assumes that you know about the syntax and parsing. So it doesn't waste any time going over that. Rather, it focuses on suggesting how to best implement/deploy it. Chapter 5 includes a nice analysis of the limitations of XML Schema and DTDs. Plus, there is something which lower level XML books often don't discuss. In those, XML examples are given using lengthy labels in tags, that have human-readable utility. Which is a strength of XML. But this comes at a cost of greater storage and processing time and bandwidth (when you transmit the XML). Given that XML is meant to be processed by software, and that humans should often only view it directly as a manual exception handling process, then having shorter tags might sometimes be acceptable, if you want to improve performance.

This is something I've encountered in my own coding, when I write/read 100Mb XML files. The sheer size of these leads me to define tags with labels of just one or two characters. Which does make them harder to manually read. But my reading routines run faster.

Erl also suggests that if you have personnel who want to learn XML, books may be far cheaper than training courses offered by third parties. Granted, this is a little self-serving, because he is saying this in his book. But no more so than asking some consulting company if you should hire them to teach XML.

Now consider the Web Services part of the book. There has been a veritable laundry list of second generation technologies developed. Like Transaction, Coordination, Security, Policy, BPEL4WS, Attachment and Addressing. Erl tries to pull these together into a coherent usage framework. The book does not go into the details to each technology. That is the purview of other books. Rather, Erl discusses integrating these into your development. Helpfully, he points out that any specific application usually only needs a subset of the above. Which is vital in learning and using them in a modular fashion. Analogous to how in java, you don't have to know all the class libraries that come with the latest java, in order to usefully program.

He offers another tip which alone may justify the entire book to you. If you asking a vendor for a full enterprise WS conversion of your legacy applications, she often gives a hub and spoke model. In this, the spokes are modifications of your applications, and the hub is written by the vendor. Typically, this gives a vendor lockin. So beware!

Reading this book I felt like the author slapped me in the face and then dunked my head in cold water all the while saying "look, this is what it's really like!". I thought I understood everything I needed to about Web services, but this book hit me with a dose of reality that really opened my eyes. It's a big book, and when it first arrived, I looked at it and thought "no way am I reading all of that", but once I got into it I didn't want to put it down. Still haven't finished it all, as some of the stuff doesn't pertain to my work, but I got thru a good 75% of it (which is a lot for someone with my attention span). The author of this book has no qualms about stating things the way they really are, as opposed to the way they should be or the way they are advertised to be. I wish there were more books like this.

This is the book to have for analysts, architects and managers working on service-oriented projects.

In contrast to other books, this one focuses on the concepts and processes involved in service oriented architectures not just on specific protocols or development platforms.

After reviewing the building blocks for service-oriented architectures, Thomas Erl goes on to offer advice on setting up teams and projects, designing services, selecting appropriate tools and aligning organizations to reap the full benefits of service-orientation.

With its easy to understand language and diagrams, this book helps technical and non-technical team members to quickly establish a common vocabulary for the elements of a service-oriented project.

The process blueprints for the analysis and design phases of a project also offer a great starting point for kick-starting the migration to a service-oriented organization.

It's the first book l recommend to my clients.