<My brain hurts/>

By Mark Gibbs
Network World, 04/22/02

As you may have noticed, over in my evil twin's domain, Gearhead, the topic of the last gazillion columns has been Web services. It has been a feast of acronyms, a smorgasbord of technology and a deluge of detail. We have been extended with XML, lathered with SOAP, schemed with XSD and rocked by RPC, not to mention directed by UDDI and serviced by WSDL. I don't know about you, but my brain hurts.

While Gearhead has taken a pretty good swing at the nuts and bolts of Web services, what Gearhead hasn't and won't be doing is covering the other half of the equation: the business issues. What will Web services do for business?

There has been too little coverage of why, as business people, we should be concerned with using this panorama of technology. After all, if you peel away the bits and bytes and take a look at Web services, there's definitely the appearance of the emperor's new clothes. Once the technology is out of the way, what do we have? Some processes that are the result of new code or legacy code being presented as XML content through services over TCP/IP connections. Doesn't sound very "new" does it?

But the opposite is true. It is all about the context of how Web services are structured, presented and delivered rather than the mechanics of doing so that makes Web services such a powerful idea and profound business strategy.

What's happening is we are taking the first small leap of the next evolutionary jump in information technology. In fact, let me be more bold than that: This evolutionary jump will not be just an IT phenomenon - it will change the way our culture functions in terms of business practices, communications, entertainment and education.

The windup for this jump lies in the explosion of XML and the lack of structure in our legacy data. I have read that in the average corporation, unstructured content accounts for some 84% of all data. While knowledge-mining tools can extract information from the data, the accuracy of that process is questionable: You will only be able to establish the context of some small percentage of the data you own and that context won't be exact - it will be assumed from inference and clues concerning the location, known sources and dating of the data.

Sure, sophisticated data mining and analysis tools are available - see my Web Applications newsletter on an interesting tool called Stratify Discovery Server. These kinds of tools can have tremendous value, but they aren't going to deal with all your legacy data. Time and expense set limits on what can be mined.

But it is the future that matters. As XML becomes as commonplace as, say, HTML, we will start to have knowledge about our business operations built into our data. We will build implicit procedural knowledge into our business processes simply by producing structured data.

Then we'll wrap that vision in defined services to control access. Digital Rights Management (DRM) and privacy controls, � la the W3C's Platform for Privacy Preferences Project (P3P), will make the distribution of data and information far more controllable than it is today. Security becomes easier because you know what you own. If you place a value on some type of data, you should be able to find out exactly how much of that data exists, where it is and who has access to it.

Now this might sound abstract, but think of it in practical terms - in terms of knowing how every piece of data was generated and why, and where it fits into your business and how it can be moved around, distributed and modified. The implications for how we do business in the private and public sectors, and how we are entertained, informed and interact are profound.

It is going to be quite an evolutionary step. I expect my head will hurt for some time.

Discuss your aching head with backspin@gibbs.com.