The Web Services Advisor
(To receive this column in your inbox,
click Edit your Profile and subscribe.).
The new year is not quite a month old, so it's a perfect time to take a look ahead at the rest of 2004. What will the year be like for Web services? Will it be the year that it finally breaks through, or will the technology percolate slowly, as it has the last few years? I surveyed the experts and here's what they have to say.
Will 2004 be a breakthrough year?
The economy has started to pick up and IT spending, for the first time in memory seems to be on the upswing as well. So in theory, anyway, that should bode well for Web services, because in tough economic times businesses shy away from investing in new technologies and those without a proven track record. Instead, they invest in technologies that can offer immediate return on investment (ROI).
That means that pickup in the economy should help boost enterprise investment in Web services, believes Jeff Mitchell, Managing Director in Integration Services for Bearing Point. "We saw that in 2003 investment was heavy in ROI, but there was little enterprise architecture work, because there is no immediate ROI on it. But we think that we'll see more spending on architecture work, and so on Web services, in 2004. And as spending loosens up, we'll see increasing amounts of it."
In particular, analysts expect there to be a big push towards service-oriented architecture (SOA), in which software components can be exposed as services on the network, and so can be re-used time and time again for different applications and purposes. In SOA, developing new applications can be a matter of mix-and-match: decide on the application that you need, find out the existing components that can help build that application, glue them all together and you're done. At the heart of SOA is Web services, the glue that holds it all together.
Mitchell says that 2004 should be a "breakout year" for SOA and Web services, a view echoed by Jason Bloomberg, senior analyst with ZapThink, who says that he saw the pickup in implementing SOAs at the end of 2003 and expects that to accelerate in 2004.
"In 2004, most companies will put together entire SOA teams; it won't just be a few isolated developers," he says. "There will be teams that cut across different departments and divisions, especially in the larger enterprises." And as SOA goes, so goes Web services.
What else will be hot
What other Web services-related technologies will become popular in the coming year?
"The next big thing is managing Web services," believes Dave Sanders, Vice President in charge of Bearing Point's Integration Services for Public Services. That means authentication, security, exception-handling -- in short, the whole suite of tools necessary for deploying and managing Web services in large-scale enterprises. To date, most Web services projects have been relatively small or pilot projects and so management has not been a top priority, because it can be done on an as-needed basis. But as corporations look to build SOAs, management of Web services becomes a much higher priority.
Related to that is how to use Web services to control business processes, via the Business Process Execution Language (BPEL) standard. BPEL choreographs Web services components so that they can more easily be strung together and work in concert to accomplish specific business purposes. They provide a standards-based approach to connect Web services to one another to create higher-level business processes than a single Web service module.
ZapThink's Bloomberg says that "BPEL is forward-looking and we expect to see more of it in 2004."
Patrick Gannon, President and CEO of the OASIS standards-setting body, says that he expects advances in anything to do with the infrastructure framework supporting Web services. So he expects that Web Services Distributed Management (WSDM) specification, which covers how to manage Web services, will become an OASIS standard in 2004 and this in turn will spur the development of software and tools for managing Web services. He expects as well that the OASIS WS-Security standard will means that security and identity management will be important Web services-related technologies in the coming year.
"We'll start to see companies showing practical demonstrations of these technologies in 2004," he says. "Mainly we'll see them being used inside the firewall. I don't expect that we'll see large-scale industry deployment outside the firewall in 2004; that's still a year or two away."
He doesn't expect that mobile Web services will see much activity in the coming year, because the standards and technologies are too immature.
But he sees e-business-related Web services to take off in a big way, after standards are developed. He says that he's seeing a great deal of interest and activity in specific, niche areas -- in other words, specific industries will develop their own Web services standards targeted only at their industries, which he says will be a catalyst to a much wider deployment of Web services in those industries.
The bottom line for the coming year
What does it all add up to? Expect a pickup in the economy to lead to greater widespread deployment of Web services. Most of the work will be architecturally related and inside the firewall or with trusted business partners instead of outside-the-firewall activities. Management, orchestration and security will all be important. So don't expect big fireworks in the coming year, although you should expect the groundwork to be laid for much greater future growth.
For related Articles and Commentary:
- Check out what are experts are predicting: What's in store for Web services '04
- Read Preston's predictions from last year: Looking forward on Web services in 2003