net.work

The Way Business Is Moving

net.work published by
Issue Date: February 2002

Web services @ your service

February 2002
Heike Beisiegel

Gone are the luxurious and happy days when Internet hype put a confident smile on everyone’s face. For many, deep frowns and a raised eyebrow soon replaced these smiles. At present, cynicism, scepticism and disillusionment are the preferred sentiments expressed by IT specialists at the mere notion of ‘a new innovation’, ‘the next wave’, ‘emerging technology’ and ‘third revolution’. But this bleak plateau is predicted to end soon when e-business becomes ‘just business’ and analysts are hoping that companies will soon step from the gutter of disenchantment onto the slope of clarification and eventually stand, head held high, on the plateau of prosperity.

Integration is touted as one of the success factors, of an effective e-business world. That is, creating a world in which computers all speak the same language and create conversations, which ultimately are crucial and profitable for businesses. The publicity, therefore, enveloping Web services is very real at the moment and with heavyweights such as HP, IBM, Microsoft, Oracle and Sun backing the concept, the model's authenticity is validated. So why should you pay attention to Web services and why do analysts such as the Meta Group believe that Web services are starting the third revolution of the Internet? The answer could lie in the user or business experience. Samir Patil and Suneel Saigal explain: "Web services not only promise consumers a more consistent and uniform experience but also allow them to integrate and personalise data and services from diverse sources. Today, for example, a consumer who wants to research a stock, trade on-line, and plan the tax strategy for the transaction might have to visit several World Wide Web sites. Web services, by contrast, would bring best-of-class offerings direct to the user's desktop. New intermediaries may integrate and deliver these services - some free, some fee-based - from a number of different vendors."*
Gartner Group once stated that during a slow economy, there is a strong financial emphasis on cash flow. This assertion is the window of opportunity for organisations to use Web services and position themselves competitively for recovery.
According to Patil and Saigal, Web services will cut the amount of time and money needed for system integration, the single biggest IT expense of most companies. Savings of up to 20% are possible, mainly through reductions in the cost of developing interfaces among systems. They go on to say that systems integration firms may not need to worry, however, since an increase in the volume of integration projects could offset the reduction in unit costs.*
Jay van Zyl of Rubico believes that Web services go beyond the initial integration promise. He says, "One of the main hurdles early e-business adopters failed to negotiate was the whole issue of applications integration. Not only could they not get that elusive supply chain collaboration systems integration right, but also their own internal systems tied together in any seamless format. IT was an expensive exercise that failed to deliver. The emergence of web services promises to change all this."
Van Zyl says that while this year is seen as the year Web services begin to move into the enterprise, Gartner Group is predicting 2004 to be the year when web services really dominate new applications at Fortune 2000 companies. But it suggests that IT should start getting ready this year.
According to a Gartner Group Research Note, entitled Web Services: 2002 and Beyond, "2002 is the year companies need to start putting together teams of developers, who will become well seasoned, to lead bigger and more valuable Web services efforts at a later stage." The Gartner Group timeline predicts Web services to emerge as "the next generation of platform middleware" in 2003, when, it predicts that 80% of platform vendors will be supporting Web services architectures.
Van Zyl sees Web services initially being used internally to facilitate integration shortfalls, but believes there is no reason, providing standards fall into place quickly, that Web services cannot go beyond the firewall to integrate partners, suppliers and customers.
"But Web services down the line are about a lot more than integration. Being components-based they open up a brand new affordable and flexible IT world where companies will need to buy and maintain only the IT technology they really need and download components for other applications on an as-needed basis as Web services," says Van Zyl. "It will need a complete shift in the way people think about building and deploying applications. In reality, in the future, it will not be about building applications, as we know them today. It will be about building digital services."
What exactly are Web services?
Michael Barnes, in his Meta research paper: Understanding Web Services, says that web services are self-contained, modular pieces of application functionality (eg, components, objects) that can be published, accessed, and invoked across a network. Unlike existing component architectures, Web service implementations (eg, HP NetAction, IBM Web Services, Microsoft.Net, Oracle Dynamic Services, Sun Open Network Environment [ONE]) are designed to support loosely coupled (eg, message-based) integration of standards-based application interfaces (eg, XML). We expect the focus on asynchronous integration to greatly reduce the costs and skills required to maintain distributed, component-based applications (versus primarily synchronous, remote-procedure-call-based integration). The acceptance of XML as the standard for exposing these services (versus platform-specific application programming interfaces [APIs]) will also diminish the costs and skills required to integrate both components within a system and components (or services) across systems.
David Setshedi, director of .NET and the developer group at Microsoft South Africa says that Web services are a new way of calling on data. He explains that Web services revolve around making data available as data points in 'the Internet cloud'. He also maintains that Web services allow you to access data on any device; and with the aid of XML users can intelligently mine data. Setshedi describes the Web services world as a business and personal world where everything is connected and where systems and processes are fully compatible and interoperable.
Stefano Mattiello, managing director of Sun Microsystems SA, describes the Web as having emerged as a versatile platform for high-value solutions, significantly increasing companies' and users' supply, access and use of services. But contrary to popular belief, the Web does not only refer to the Internet, it describes an integrated network of various technologies.
He says that one of the most significant developments regarding the Web is the emerging role of smart Web services. Sun Microsystems describes today's smart Web services as "Services on Demand".
"We prefer calling it services on demand as the concept is really the foundation for a modular, flexible and automated access to digital assets, including computer resources, from virtually anywhere," explains Stefano Mattiello, managing director of Sun Microsystems SA.
"Services on demand offer businesses a way to improve efficiencies in business software usage, not only delivering value and productivity today but also preparing a business' IT infrastructure for tomorrow. Importantly, the model promises better cross-business integration, improved efficiency and closer customer relationships."
Mattiello sees this an evolution and not a revolution. "For this model to be attractive to business, companies must be able to leverage existing application assets and expose them as services. Rather than connecting to or integrating with existing resources, Services on demand can leverage and extend them.
"Services on demand should be the heart of a business's profitability and competitiveness as the model enables many benefits, while providing a smooth, clear path to a computing platform that will serve an enterprise's needs for years to come," he concludes.
According to Mark Gower, marketing director at Oracle SA, the business world has made a monumental investment in the Internet over the past few years, and 2002 will see pressure being brought to bear on those organisations to deliver a return on their technology investments. He says that Web services are a key way of realising that return. He also sees Web services as playing a huge role in future software development. He believes companies should undergo the following, to get ready for Web services: "Companies ideally should make a fundamental mindshift away from server-based business applications to Web-based applications, whether they host them in-house or use an external hosted service provider. They should look at their business and decide which business processes could be best deployed over the Internet."
International analysts are quoted as saying that 2002 will be the year of Web services. The general consensus in South Africa is that locally this is not true. Darryl Frost the software architect at IBM SA's Software group says, "Although some of the standards are starting to mature, there are still some issues that will hamper the adoption of Web services, for example the issue of security and a reliable transport layer. In some cases from a client view there needs to be several enterprises offering the same service. An insurance premium quote service, for example, is only useful to the broker community if offered by several companies. Another issue revolves around the capacity of organisations to adopt new technologies. No, I do not believe 2002 will be the year of Web services; however there will be an increase in rate of adoption of the technology. What businesses need to come to grips with is developing new business models that exploit the technology. Businesses by the same token need to start putting plans in place for evaluating the use of Web services.
The first practical applications will be inside organisations for process integration.
Traditionally financial services and airlines are early adopters of new technologies and I see this trend continuing. As to what the applications will be, I can only guess.
Although the real value of web services will be realised from inter enterprise communication, companies can start experimenting with Web services in order to integrate intra company processes. This will allow them to get familiar with the technologies before exposing services to parties external to the organisation.
If processes within the organisation are not currently tightly integrated, in order to fully realise the benefits of Web services, organisations will need to address these integration issues. This is a good starting point for using Web services within an organisation, integrating existing processes within the organisation."
Kevin Barnard, regional program manager for UNIX servers at HP South Africa, says that Web services is not a new terminology yet; it is just a reiteration of the age-old idea that software should be reusable. He says that for Web services to be used at their optimum, the marketplace needs to be educated as computers are still seen as impersonal. He believes that South Africans are still 'touchy-feely' and people-driven, as opposed to their European counterparts. He asks, "How many people are actually using Web services?"
Getting ready
To get ready and transform your business into a Web service architecture, Patil and Saigal ask the following questions:
* How can you use Web services to reduce information technology costs and time to deliver?

* What new revenue-generating services can your company offer via the World Wide Web?

* What are the new ways to interact with your trading partners? What threats and opportunities will emerge?

* Do you have the organisational and IT capabilities to operate in the Web services world?

* How will each of the competing Web services architectures (for example, Microsoft.NET vs SunONE) compare in terms of costs, capabilities, and your company's IT environment (Windows, Unix, and so forth)?
*
Service with a smile
As a business-to-business enabler, Web services have the ability to create a strong value proposition. Web services will speed up the business-to-business e-business model. As Frost sums up, "What e-business did for business-to-consumer, Web services will do for business-to-business. It will make process integration simpler."
* The McKinsey Quarterly, 2002 Number 1
Web services terminology
SIMPLE OBJECT ACCESS PROTOCOL (SOAP)
Developed by Microsoft, IBM and others - and submitted to the W3C as an open standard - SOAP outlines how the envelope for XML messages should be written and how the messages themselves should be encoded.
WEB SERVICES DESCRIPTION LANGUAGE (WSDL)
Developers use this to expose what a Web service can do. When a client or another Web service makes contact with a Web service, the initial response comes in the form of a WSDL message.
UNIVERSAL DESCRIPTION, DISCOVERY AND INTEGRATION (UDDI)
Think of UDDI as a directory for Web services. It has three parts: a White Pages that lists contact information about the company that developed the Web service; a Yellow Pages that lists Web services by such categories as geography and industry code; and a Green pages that includes WSDL descriptions, e-business rules and how to invoke the service. The idea is that, eventually, every public Web service will be registered in a grand UDDI directory, so Web services can dynamically discover each other, hook up and integrate without human intervention.
E-BUSINESS EXTENSIBLE MARKUP LANGUAGE (EBXML)
Jointly sponsored by a UN standards agency and Oasis, a non-profit consortium, ebXML purports to create a complete framework for XML-based B2B e-commerce, particularly those involving secure transactions. Industry players disagree whether the adoption of ebXML is necessary, or whether existing Web security protocols will suffice. (Source: Make Way for Web Services, CIO Magazine)
Vendor visions converge (but implementations will vary widely)
Microsoft and IBM are the two infrastructure vendors that have done the most to define standards for Web services implementations. However, all major infrastructure vendors have announced plans to support Web services (including the aforementioned core standards) within their respective software stack. Sun (via ONE) and BEA (via BEA WebLogic E-Business Platform) have both laid out their own Web services visions.
HP was the first vendor to articulate a Web services vision (via the original rollout of e-speak several years ago) but could not deliver on this vision due to a lack of focus and a lack of the underlying technology components/software infrastructure required to enable Web services. The acquisition of Bluestone late last year provided the core J2EE application server (now part of the NetAction suite) necessary to make E-Speak (a Java-based Web services implementation that was originally designed as a tightly coupled, distributed object implementation, but has evolved toward a loosely coupled implementation that supports XML, SOAP and UDDI as underlying standards) more than a vision. However, though HP appears heavily committed to supporting Web services via NetAction, the company is a late entrant into the J2EE application server market and will therefore have trouble keeping pace with far better established vendors (eg BEA, IBM).
Unlike HP, Oracle was relatively late in announcing its Web services strategy (Oracle9i Dynamic Services was announced in December 2000) and has focused primarily on a database-centric implementation (versus the application-server-centric implementation targeted by the other major infrastructure vendors). Much like the company's J2EE development and deployment solutions (eg Oracle9iAS), we expect Oracle9i Dynamic Services to be a compelling option for large Oracle shops interested in implementing a Java-based Web services strategy). However, this approach will have limited appeal within the broader market.
Business impact: A standards-based, technology-neutral means of connecting systems will dramatically reduce the cost and complexity of automating business relationships.
Bottom line: Organisations should begin to work with Web services to ease the implementation of B2B and enterprise application integration systems. Strategies should assume the extensive use of Web services by 2004. (Source: META Group research: Designing Web Services by Daniel Sholler)
XML programming language is crucial for systems integration
It is crucial for South African businesses to realise the value and importance of using XML and its native variant in their Internet applications and development. Quite simply put XML means data anywhere. To the businessperson, XML and Native XML might sound like 'techie-speak' but in fact it is more a business issue than a technical one, says Mimmo Mattiello, principal sales consultant - tools for Oracle SA.
The reason is simple. Today's enterprise relational databases offer business-critical features such as flexible data access, lightning-fast query speeds over 'galaxies' of data, and integrated support for multimedia data and data warehousing to support personalisation on a massive scale.
They offer the reliability, availability, and scalability that the modern 'dot-com' cannot exist without.
"However, even with full-featured, high-performance relational databases at the heart of their Internet architectures, e-commerce sites claim their key development challenges continue to be integration, integration, and integration," says Mattiello.
"Integration with back-end ERP and CRM systems from multiple vendors. Integration with systems from partners in their supply chain, and integration with data warehouses for unprecedented levels of personalisation."
Leading relational database vendors realised early how powerful a technology XML would be in providing solutions for these kinds of information and application integration challenges and have been working at a feverish pace to tightly integrate XML support into their database platforms. But the debate regarding XML versus Native XML is irrelevant says Mattiello, who believes that the two technologies are integral in a database solution.
"The difference between XML and Native XML is that XML deals with data-centric documents while Native XML deals more with the content centric document and, quite obviously, it would benefit any organisation to have both technologies in the same database, talking to each other."
While the term 'Native XML database' first gained prominence in the marketing campaign for Tamino, a Native XML database from Software AG it has never had a formal technical definition.
Beyond simply supporting the XML standard, Oracle has begun to participate actively in steering the standards to directly influence the evolution of XML.
At present, Oracle is working together with IBM, Microsoft, and others in the W3C XML Working Group to define an Internet Standard for 'XML schemes'. This effort will create a standard to describe the structure and datatypes of a document's elements to enable more automated and seamless integration of XML with databases and programming languages in the future.


Others who read this also read these articles

Search Site





Subscribe

Previous Issues