If I were a lot more paranoid than I am (well, perhaps at least a little more than I am), I might suspect that the various free e-mail programs were a social engineering attempt by the big software coalition (yes, I know, it doesn't exist) to ultimately change the way we use our computers.

What does this have to do with service-oriented architecture? I'll get there, don't worry. But first, as to why would this be social engineering, let's take a look at the concept of Software as a Service (SaaS, for acronym fans). Software as a Service is not an entirely new concept, but it's one that has always faced an uphill battle. Previously, and even now to a certain extent, continuous connectivity was a serious issue. Reliance on software provided across the Internet can be problematic if the connection is slow or interruptible. While broadband access has solved this problem for many, there are still many challenges to be faced with software that doesn't reside on your local hard disk.

This challenge is something that faces both consumers and businesses - while a business may be able to secure faster, more reliable communication through redundant network access, slow response time can be a fact of life in such a scenario. Even then, there's no guarantee that an upstream DNS server may not be under attack or down. In the case where the software is local, this is not a problem, and might even increase productivity slightly by preventing employees from surfing the Net for a period of time.

However, when the software sits in cyberspace (say that three times quick), it becomes a different matter. Downtime costs money, as do delays. I've heard numbers relating to call centers that say saving one second of call time results in a savings of a million dollars over a year. So, turning that around, delays introduced due to network latency could cost millions of dollars - not something a thin margin business wants to consider.

So social engineering comes into play. Free services, such as e-mail, start to address some aspects of this issue. First of all, with steady broadband connections, we can see that by and large, the service provided is acceptable. Not perfect, I think we'd all admit - and in many cases the interfaces provided are inferior to those of a thick client e-mail application. But, for the most part, you get reliable e-mail access, as well as reasonable functionality, without having to have a home e-mail server or any installed software. You begin to see that Software as a Service is possible.

There are other examples of services that aren't free, such as salesforce.com, that are successful in providing Software as a Service. Depending on your definition of software and of service, we could identify many more.

Service-oriented architecture, founded on the principle of loose coupling and the distinction between a service and an application, is the perfect driver for Software as a Service. As organizations begin to decouple their services from the silos that an application focus has placed them in, they begin to recognize the intrinsic value of the service and its relationship to their business processes. The value of an application does not become less, but the nature of an application changes to one where services form the core of the undertaking, and the presentation aspects of the application become the unique value.

With that in mind, it becomes easier for Software as a Service to become a reality. An organization already making use of services finds it simple enough to integrate an external service. The reverse is also true. An organization using services finds it easier to expose its own services to external customers - in essence to take its services to market by providing Software as a Service.