A Roadmap for the Enterprise
First edition; 355 pages
ISBN: 0-13-009719-5
( by: Alex Nghiem )
Preface
Introduction
Today, more than ever, enterprises are faced with great challenges: rising development costs, budget cutbacks, and increasing customer demands, to name a few. Decision makers must determine which projects to fund and are often confronted with an existing technology infrastructure that already includes a hodgepodge of technologies, such as object-oriented technologies, enterprise application integration (EAI), enterprise resource planning (ERP), and customer relationship management (CRM).
Against this backdrop, Web Services emerge promising many wonderful benefits, such as reduced integration costs and a low learning curve. How can a technology planner reconcile these promises with conflicting real-life experiences involving incomplete standards, lack of security features, and other key issues? How should these gaps be addressed? Do Web Services replace existing technologies or do they augment them? Has any firm adopted Web Services successfully? If so, what were the benefits and how was it done? What makes Web Serv-ices different from previous attempts of interoperability such as Common Object Request Broker Architecture (CORBA)? And why is it inevitable that Web Services or the like will be adopted on a grand scale?
The answer is that early adopters of Web Services have indeed used the technology successfully to achieve a variety of goals, including providing better customer service (Putnam Lovell Securities), building a digital marketplace with reduced development costs (Pantechnik International), and saving the adopting firm millions of dollars by building a sophisticated procurement platform (Talaris).
An enterprise that is seriously considering using Web Services needs to identify what gaps are relevant and how to address these gaps. Many of the gaps in the standards are rapidly being addressed by emerging technologies or third parties such as Web Services networks.
Understandably, many organizations are reluctant or even nervous about investing in Web Services for the long haul, having lived through too many rosy predictions. With amazing consistency, most of the firms interviewed for this book expressed the concern that Web Serv-ices will indeed be ubiquitous for a variety of reasons:
- They are built on existing infrastructure and open standards (Hypertext Transfer Protocol HTTP, Extensible Markup Language XML, etc.), which means many firms can launch pilot projects without a huge initial investment. Standards are important not because they are cutting edge—they are important only if they are widely adopted. Web Services are a social phenomenon rather than a technological one (elaborated further below).
- Beyond adopting open standards, vendors are actively working together to ensure interoperability between their technologies. In addition to independent standards organizations such as the Worldwide Web Consortium (W3C), vendors have formed interoperability groups to produce compliance tests. An example of such a group is the Web Services Interoperability Group.
- Every major vendor has announced support for Web Services. One technology executive pointed out that a firm would have to actively try to keep Web Services out because the world’s largest software company (Microsoft) has embedded this functionality in its Office products.
The best-selling book The Innovator’s Dilemma coined the term disruptive technology to refer to a technology that, once introduced, has the potential to dramatically affect the equilibrium of a market. A disruptive technology has three characteristics:
- It can pull a large population of less skilled or less wealthy customers into the market.
- It can get traction only if it helps people accomplish something they are already doing in a better or faster way.
- It can gain a foothold without being as good as an established product.
Let’s explore whether Web Services have these characteristics.
Currently, many firms are dealing with high integration costs on too many projects-they have no choice but to buy high-end packages for performing even simple integration. In addition to being expensive, many of these packages require scarce high-end skills. There are entire market segments (small- to medium-sized businesses, predominantly) that cannot afford these types of initiatives.
While they are not appropriate for all integration projects, Web Serv-ices can be used in many scenarios that, until now, require a high-end integration package (such as an integration broker or EAI). A consistent message in the industry is that Web Services can democratize integration (point 1 above) and make the benefits available to a much larger audience due to a significant drop in costs, thus resolving the criticism in point 1 from the previous list.
To address point 2, firms are already performing integration and, in many cases, are looking for better methods. Again, broadly speaking, integration takes on many forms and involves exchanging information between organizations. Thus, integration can then include transferring files using file transfer protocol (FTP) or email, integrating two back-end systems via EAI, and the like. While Web Services will not replace EAI in the near future, they can automate and streamline many integration scenarios where firms are performing manual tasks. You might be surprised to learn that many enterprises are still exchanging very critical data through a manual FTP process.
The third point is critical to understanding the rate of adoption: many technologists would argue that Web Services are not as capable as existing technologies (EAI, middleware, etc.) and should not be considered. A disruptive technology does not need to compete with existing technologies; it needs to be good enough to address the needs of a market segment that is suffering from a business pain. Again, not all integration problems require a high-end integration package or middleware; in those cases, Web Services provide a compelling alternative.
As an analogy of a disruptive technology, consider the introduction of the personal computer (PC) almost 25 years ago. At the time, many incumbents (including IBM and DEC, who dominated the mainframe and minicomputer market, respectively) did not consider the PC a threat because they could not envision why a business would use such a limited tool. Because the businesses using computers at that time included banks, airlines, the military, and the like, incumbents thought computers needed industrial-strength storage, high throughput, and rock-solid reliability. However, they overlooked a large market: people such as analysts and business owners who performed financial modeling tasks on a daily basis. People like this met all three characteristics of disruptive technology in the earlier list: a large audience of less skilled and less wealthy customers who were already performing their tasks routinely and who did not need a product that competed with the one offered by the market leaders.
Fast-forward 25 years to the present and witness how the entire industry has changed. Companies that were barely formed (Intel and Microsoft) are now among the world’s most valuable companies, and one of the two then-market leaders (DEC) is not even around. Many analysts and technology executives are predicting that the introduction of Web Services can have an even bigger impact on the industry than did the introduction of the PC.
Using this book as a roadmap, you should be able to navigate through many of these issues to decide how Web Services should fit into your enterprise’s strategy and whether the business benefits are compelling enough to launch a pilot project. At that point, there are many other wonderful references on how to implement Web Services.
Audience
The purpose of the book is to demonstrate the business benefits of Web Services for organizations of all sizes, along with providing the supporting technical background. It attempts to remove much of the noise and hype surrounding this technology and provides a foundation on which to make informed decisions.
It is first and foremost intended as a roadmap for technology planner roles such as chief information or technology officers (CIOs or CTOs), directors of technology or product strategy and software architects. These are the people who are responsible for determining whether this potentially disruptive technology will replace or complement traditional integration technologies such as middleware and EAI.
Others who can benefit from the book include business analysts, who have to decide on whether the claimed benefits—sometimes outrageous ones—provided by Web Services are compelling enough to justify funding a pilot project, as well as investors who are considering whether this market is worth taking a second look at (and investing in).
Web Services are evolving so fast that it is almost impossible to predict where they are headed. Their benefits are illustrated through many extensive one-on-one interviews with executives at leading technology firms and by looking at real-world case studies of early adopters. It is the executives who are making the tactical and strategic decisions regarding technology who will, in many ways, determine the rollout of Web Services. Through the unedited detailed sessions with these executives, you will have a 360-degree view of the marketplace. Then you can reach your own conclusions.
Structure
Chapter 1 starts with some of the business conditions that drive integration at the enterprise level and then provides an inventory of technologies that currently exist to address some of these issues. By discussing the challenges that are present in adopting these technologies (object-oriented technologies, component-based development, EAI, etc.), the chapter illustrates how Web Services came to be.
Chapter 2 discusses what is widely known as the basic Web Services stack, including XML, Service-Oriented Access Protocol (SOAP) (historically called Simple Object Access Protocol), Web Services Definition Language (WSDL), and Universal Description, Discovery, and Integration (UDDI). It then concludes with a discussion on how Web Services will augment (and in some cases replace) the technologies discussed in Chapter 1 and how they will be adopted in multiple phases.
Chapter 3 covers Web Services at the enterprise level and focuses on a key area: messaging. Some key decisions that need to be decided when designing Web Services are whether to use synchronous vs. asynchronous architectures and whether to use a point-to-point or a publish-subscribe model. The pros and cons of these decisions are discussed in detail.
Chapter 4 focuses on two key Web Services platforms: Java 2 Enterprise Edition (J2EE) and Microsoft .NET. Ultimately, Web Services have to be deployed in one of these two platforms. This chapter discusses how the two platforms approach Web Services. It also includes a high-level discussion on how the various J2EE licensees (IBM, BEA Systems, Iona, etc.) differ in their implementations.
Chapter 5 discusses two key areas of concern for large-scale Web Serv-ices deployment: security and scalability. The security standards are not yet defined, but the W3C (the international standards body) is evaluating many mechanisms including digital signatures and encryption. The chapter then concludes with a discussion on how to architect Web Services for scalability.
Chapter 6 discusses an emerging market called Web Services networks. These firms provide value-added services including provisioning, guaranteed messaging, and centralized reporting. Two vendors, Flamenco Networks and Grand Central Communications, are profiled, and the chief executive officers (CEOs) of both are interviewed to gain their perspectives on how Web Services will be adopted.
Chapter 7 lists the common architectural patterns that should be considered when adopting Web Services. Each pattern is described with the necessary preconditions for adopting it along with the pros and cons.
Chapter 8 pulls it all together and provides a high-level plan for an organization adopting Web Services.
Chapter 9 discusses an emerging trend that is often confused with Web Services: Software as a Service (SAAS). The chapter discusses the pros and cons of adopting such a solution and then provides a one-on-one interview with an SAAS pioneer, Employease.
Appendix A discusses other initiatives including ebXML, Web Services Flow Language (WSFL), and XLANG. A worldwide initiative to provide a standards-based platform to facilitate e-commerce, ebXML includes a catalog of predefined business processes that can either be adopted as is or extended. WSFL and XLANG are emerging but conflicting technologies to address workflow.
Appendix B provides multiple detailed case studies on how enterprises are adopting Web Services to lower integration costs and open new markets. Each of these firms shares the lessons it learned as an early adopter.
Appendix C provides multiple perspectives drawn from detailed interviews with many leading Web Services vendors and startups including SilverStream Software, Iona Technologies, Cape Clear, and Collaxa. Each firm provides some unusual twist in its approach to Web Services.
Appendix D is a detailed product review of the bundled product XMLBus.
Bundled Software
The bundled CD contains a one-year license to Iona’s E2A Web Services Integration Platform: XMLBus Edition. Follow the directions on the CD to install and register the product.
Continued Web Support
Addenda to the book and follow-up case studies can be found at www.bluesamba.com/webservices
