Sometimes it seems like it takes forever for new technology to be adopted, and even when a technology goes mainstream, it seems as if people cling to the old ways long after a better way has been shown. Heck, I drive a car but still have a couple of horses - sometimes you just want the old and familiar around.

Most of us are familiar with the famous Gartner Hype curve as well, which describes the adoption of a new technology in phases, from the build up of talk about a new standard (the start of the hype), through early users to widespread adoption. I suppose the steady state shown at the end must eventually decline as technologies are made obsolete, but I've never really heard of the opposite - the technology replacement curve, as it were. Sometimes generations have to pass before the technology can be put out to pasture - think about how much COBOL code is still around. What can be interesting is - as a technology reaches its sunset years and the number of people who know it decreases, there's a brief flare at the end of its life where the expertise is hard to find and once again people are paying high rates for talent in what is now an obsolete technology.

Sometimes, a technology seems like a logical, natural fit, but for one reason or another, it doesn't gain traction immediately, even though it makes perfect sense. Part of this is due to the nature of our information technology environments. They're not green fields, waiting to be developed. No, they're more like housing developments, some of which are new, some which are rundown and blighted. A new technology is like a developer wanting to clear an existing track of houses to put up new condos - sometimes the people in the existing houses don't want to sell.

Service management, including Business Process Management, is experiencing just that sort of adoption lag. Existing approaches to the problems addressed by BPM and service management are the low rent district of the IT world. BPM represents the new waterfront condo with the football stadium. Something big, shiny, and new, but with a lot of details to work out, and a lot of political maneuvering to make it actually happen.

Business Process Management provides the ability to model and implement business processes composed from services - a clearly desirable feature in the world of service-oriented architecture. It makes it possible to orchestrate the complexities of an end-to-end process through the myriad of systems that support a business. While that is undoubtedly a good thing, corporations have been making do without it for decades.

At first the technology wasn't available. Then, applications were developed in isolation, requiring manual intervention and annoying workarounds to actually implement a business process. But by and large organizations learned to cope with the non-integrated nature of their business applications.

BPM promises to change that, on top of a service-oriented architecture framework. The challenge is that we're talking large-scale replacement in order to fully achieve the vision.

Short term, we're seeing an approach to stealthy adoption that is common in the software industry - bundling the capability into other products until it becomes ubiquitous. This approach has already worked wonders for service-oriented architecture. As more and more package vendors adopted Web services as their basis for communication, the barriers to entry dropped and it became easier to use SOA then to avoid it. BPM is making inroads by being bundled into packages as well, and by targeting subsets of the market, like service management and service monitoring. As companies begin to use those tools to manage their environments, and learn the power of a process-driven infrastructure, the adoption of BPM will finally get over that hump in the hype curve.