One of the persistent knocks against IT is that, depending on who you ask, as much as US$7 out of every $10 dollars spent on software goes into installing and integrating the software once it's purchased. This always leaves the people who buy the software feeling a little burned, and frankly, gives the industry a bad reputation among businesspeople who typically fund these projects.
Despite this state of affairs, integration remains the primary activity that actually draws value out of software. A line-of-business software application that is unable to draw data from other applications will never justify its existence by creating a visible ROI to the business. This is why even in a down economy companies specializing in enterprise application integration are doing reasonably well.
The good news is that new technology is about to make a difference in this space in the form of Web services. As a concept, Web service is not all that new. Basically, it consists of network resident applications made up of components that can be called over the Net using a set of defined interfaces. And most of us have been talking about this type of software architecture for more than 20 years.
What's new is the advent of XML, which in its purest form is a self-describing data-neutral format. The key phrase in that description is the term self-describing, which means that one piece of software can automatically discover and understand the function that another piece of software can perform.
We're still in the early days of developing all the standards around Web services, but once they are in place, Web services will significantly alter how we deploy, manage, and acquire software -- and more importantly, reduce a lot of the soft hidden costs associated with enterprise software.
As does all substantial change, this will take time to develop. But we here at InfoWorld think that the implications of Web services are so profound that we've decided to devote an entire two-day conference to the topic this October in San Francisco. Details about this conference can be found at www.infoworld.com.
In the meantime, there is a war on between Microsoft Corp. and its vision of XML as implemented in its .NET architecture and the backers of Java, who tend to view XML as a complementary set of technologies that will make Java applications more accessible.
It's unlikely that either side will vanquish the other, so we can expect to see these two architectures co-exist in the market. Nor do we necessarily have to wait for the big guns in the industry to finish their architectures to take advantage of Web services today. A number of smaller companies, including NetObjects Inc., Cape Clear Inc., SilverStream Software Inc., Avinon Inc., and WebCollage Inc., all have viable offerings available today.
Overall this is a pretty exciting time in the world of software. We are on the cusp of realizing age-old visions of how software truly should be constructed. And more important, we are getting close to resolving an issue that has always plagued the industry as a whole: how to derive incremental value from software without incurring exorbitant costs.
Web services may not be the silver bullet, but it will sure go a long way toward improving what is today an abysmal state of affairs.