Customer Reviews

Web Services: Theory and Practice

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

 

 

This book is presented as a comprehensive guide to all aspects of Web services, and one that emphasizes the practical issues involved in its use. It is written for a reader who is relatively new to the subject, and therefore does not go into the minute details of it. The author expresses extreme confidence that Web services will continue to rise in importance and will even rescue the IT industry from its current slump. It certainly has had an impact in business applications, but only time will tell whether it will dominate this environment in years to come. Web services has certainly had its critics, who frequently accuse it of being overly hyped and for falsely raising expectations, as well as needing further developments in standardization before being deployed at a large scale. Indeed, the Web services standards process itself has shown signs of fragmentation, with organizations such as Liberty Alliance, Oasis, W3C, and WS-I all competing for the honor of presiding over the standardization process. In addition, vendors of Web services are already at odds with each other, some of these involving licensing and usage restrictions. This has taken place even though Web services are supposed to be a technology that is free to anyone.

Web services are defined in the book as `modular, self-contained application logic' that is developed according to a set of open standards, which the author takes to be the W3C (World Web Consortium). The other standardization efforts mentioned in the last paragraph are not therefore considered in this book. Extensible Markup Language (XML) is of course presented as the underlying basis for Web services. Web services are thus described as a `remote invocation mechanism' that is always realized using XML documents. The protocol for realizing this remote procedure call is called Simple Object Access Protocol (SOAP) and is basically an XML-based messaging system. The author describes how Web services began as a `program-to-program' solution rather than a `human-to-program' one, thus vitiating the need for it to be integrated into a GUI environment. This was changed just a few years ago, he explains, by the advent of Web Services for Remote Portals (WSRP), and a related specification called Web Services for Interactive Applications (WSIA), which expressed the need for a GUI environment in order to make the deployment of services within portals much easier.

After an overview of XML and WSDL in chapter 2, the author discusses Microsoft Web services in chapter 3. After a brief discussion of the history behind Microsoft's involvement in Web services, and the tension between IBM (Java) Web services and that of the .NET approach of Microsoft, the author moves on to a general discussion of the latter. Although there are more thorough treatments of .NET in the literature, the discussion is fairly informative. Performance issues are also briefly discussed in the context of Web services deployed on Windows servers.

The author spends an entire chapter on UDDI and one on SOAP, which give the reader ample information on these two `building blocks' for Web services. Also included is a chapter on Java and Web services, which because of the nature of Java to be `cross platform' seems like a natural language to use. The author though points out the difference in platform independence in Web services, namely that different Web services can run on different platforms, and platform independence with Java, which means that the same Web service can be ported to different platforms.

One area that could be very important in the future use and development of Web services, but is not mentioned in this book, is artificial intelligence, such as current research efforts in the Semantic Web and research in intelligent agents in networks. The goal of the latter is to manage networks without the need of humans, or at least to make their use minimal. Researchers who work in this area have expressed skepticism as to the ability of WSDL, SOAP, and UDDI (Universal Description, Discovery, and Integration) to achieve complete automation and interoperability. These would have to be altered in order to support automated reasoning. The integration of intelligent agents with Web services is will allow the agents to reason about and coordinate services over the Web. Since this will typically involve working over domains or environments that are not known a priori by the agent. This will either entail that the agents adapt to these novel environments or that the Web services themselves change so as to not be as immutable as they currently are. Whatever the case may be, the connection of artificial intelligence with Web services is one that is being currently explored and may prove to be very fruitful for both fields.

From the preface of this book: "This book is not meant for an overly technical readership looking for a 'bits-and-bytes' guide to developing or using web services. This is not a guide for writing web services software." The preface goes on to describe that the book is written for uninitiated executives (my phrase) and others who want to get a handle on what web services is all about.

This book doesn't delve deeply enough for those interested in developing or exploiting web services, but is extremely technical in many other respects. I think this book is destined to disappoint a lot of readers - it falls short of technical implementation details needed by programmers, and will probably be too technical for business executives wanting to understand more in general terms about the topic. Project managers who are considering implementing web services may find the book useful, but I'd recommend browsing it first.

I found that the writing style of this particular book was turgid and difficult to follow. It was riddled throughout with spelling errors and cliche, filler phrases and poorly-conceived charts and diagrams. A good proofreader and editor could have trimmed at least 100 pages out of this book.

Web services are really a fusion of several technologies, including XML and a host of possible programming languages. If you are comfortable with a programming language such as JAVA, *.NET, C++, or even PHP, you may want to pursue a book written about web services specific to the programming language you intend to use. Beyond this, you will want to find a book that clearly addresses a few other key topics related to web services:

Universal Description Discovery and Integration (UDDI) - Can be considered the white/yellow pages of web services.
Standard Object Access Protocol (SOAP) - Currently, the common method for invoking a web service and retrieving the results.
Web Service Description Language (WSDL) - A standardized method for describing a web service, its input paramaters and returned values, using XML.

Guruge gives us a nice explanation of Web Services that does not try to drown you in an acronym soup. But the distinctive aspect of his book is in one chapter, where he devotes special attention to Microsoft's efforts in this field. Guruge points out that through its dominance on the desktop, and with its Office applications, which can now import and export in XML format, Microsoft is by default playing a pivotal part in the development of Web Services.

The chapter is not so much about the lower level technical details of Microsoft's foray into Web Services. Instead, it is a valuable comparison with other players, especially IBM with its WebSphere approach. The chapter contrasts Microsoft's .NET Web Services with those promulgated by IBM and others. This section of the book is the most rewarding, if you are interested in a higher level view of where Web Services might be going. Other chapters are more-or-less standard descriptions of Web Services that can be found in other texts.

Guruge also talks of the possible dangers of rogue Services. Here, however, he slips a little. The data passed between Services is in XML format. Thus, he argues, there is little risk of a Service getting a malware binary. But actually there is indeed this risk. Web Services are still young, and almost invariably do exchange "passive" XML data. But binary data can be transformed/encoded in an XML format and passed along, where the recipient Service "knows" that it should decode this into a binary and then run it, after some suitable safety tests. If Web Services do take off, there will be some Services doing just this. And it is this which will be a danger point for malware to be injected.

This book does not say anything new that other web services books do not.