Hmm..according to my latest Monthly Hype magazine, Web services open a whole new vista of applications to the world. Applications that have been unseen or unimagined by mankind. Everyone is asking about the Web Services Killer App. What is the Killer App?

The Web Services Killer App was unseen or unimagined by mankind until Tuesday, February 6, 2001, at 3:42 a.m., when I dreamed that I saw a great light. I walked to the light and I clearly saw SOAP and WSDL and UDDI and a few fuzzy things like security and transaction handling all dancing in a mist.

Suddenly the mist cleared and I saw a multitude of systems working together as if it were all part of a seamless machine. Then I awoke and realized I knew what the Killer App was.

What is the Killer App for Web services? What makes them pervasive? What will make a lot of money for a lot of vendors and consultants?

First things first.

The Killer App is integration.

"What?" the shocked audience gasps. "Hey, we want our money back. Integration? What do you mean? That's what Web services is about, but what's the Killer App? It can't be just plain old integration, can it?"

My position is that, yes, integration is the Killer App for Web services. It's that simple. I know it's not as exciting as Lotus 1-2-3 or e-mail or Doom, but that's what it is.

Big systems talking to other big systems - within and across business boundaries.

I know what you're saying, "But Steve, I could have guessed that!" Yes, you could have, but I jumped the line and told you before you could tell me.

The Killer App is the ability to actually build and integrate our software without fear.

Sending XML over HTTP is a pretty simple concept. Many companies are doing that today and Web services adds a standards layer, so we're no longer one-offing the effort. It's so simple in concept compared to other technologies that it's easy to dismiss its full impact.

Many times we see examples of technology looking for a problem. In the case of Web services, it's a pervasive problem latching onto a technology. The pervasive problem is integration.

Web services standards need to mature and to continue to grow-and they will. However, for day-in-day-out developers and IT managers who are trying to satisfy the needs of the businesses they serve, Web services deliver on a promise: the promise that you can separate technical implementation from business intent.

Normally, when someone builds an integration these days it tends to be specific. It's either specific to a particular integration project - I need to hook System A to System B - or it's an attempt to create a generalized XML integration interface that the rest of the world can use. But it's not based on standard, so it's specific to a given custom protocol. Clearly, one-off integration projects are time-consuming and costly.

Every company on the planet suffers from integration problems. We'd all love to have a clean enterprise architecture. But we don't. Mergers. Acquisitions. Previous CIOs. Systems that were sold because the CEO's nephew was the salesperson. Divisions purchasing their own solutions to circumvent IT. Changing technologies. These have left us with a hodgepodge of different technologies in our organizations - and no one wants to pay to have these systems rewritten. Especially in today's economy.

So being able to integrate our systems internally adds value. Big time. And being able to create business processes across business boundaries adds even more value. It's easier to do ROI and cost-justification on the internal integration today because going outside the business means thinking outside the box - and having business partners who are like-minded. It will come, but it will take some time to get everyone on the same page.

So what is the Killer App?

It's having your Java application server be a Web services front-end for your mainframe and EDI transactions. Having a portal or JSP applications be able to use Web services calls to those transactions to create thin-client applications with straight-through processing to the backend. And then writing a Visual Basic application that calls those same Web services to create a client application, so your customer service people have an application with a rich GUI. And six months from now, allowing wireless devices to call those same Web services. And a year from now, having your business partners look up your services in a UDDI directory and call them. This is the kind of integration I'm talking about when I say it's the Killer App.

Being able to use the tools, languages, frameworks, app servers, data formats, and platforms of your choice without fear of massive integration issues is pretty killer to me - and to IT organizations worldwide.