Enterprise IT has split into two lanes. In the slow and steady lane, we have Core IT. Core IT is the traditional work of IT departments. It maintains systems that support the work of hundreds, thousands, or tens of thousands of people. They need systems that work reliably and have an assumption of security.

I got the "Core IT" term from Sacha Labourey of CloudBees, who was tactful enough not to call it "Slow IT." But Core IT is slow by design. It has strict requirements that slow it down. It is risk averse. It is usually funded with an annual budget, which is the way an organization tells its IT department to take at least a year to do anything.

In the other lane, we have Fast IT. Fast IT is the stuff that doesn't fit into an annual budget because it must be delivered fast and changed quickly.

Fast IT can take many forms, including:

• Mobile apps
• Marketing and sales sites and campaigns
• Everything that customers and partners are doing with your data and your API's
• All of the new products, software and systems that your own people are moving to public cloud infrastructure, where they can do new things with Continuous Agile.

Investment in Fast IT and SaaS is growing faster than the investment in Core IT. Through this process, many Core IT functions will eventually migrate into Fast IT systems on public clouds. This is the vision of "Web Scale IT."

Core IT is often built with a top-down engineering method. In the engineering method, we know our requirements. After we list out the requirements, we can engineer a system that meets most of the requirements. Then, we build the system.

Fast IT is built with the scientific method - experimentation. We don't know what will work, or we don't know how to make it work, so we do experiments to find out. We use Continuous Agile to keep this process moving, and we pay a lot of attention to measurable results.

I once read a study that compared engineers with scientists. It found that engineers often have extremely specialized knowledge about specific systems, and they have to stay inside their home organizations to find people who understand what they are talking about. Core IT fits this engineering model with a local buildup of specialized expertise. Scientists work on more general problems, and they spend a lot of time communicating globally and sharing ideas. Fast IT fits this scientific model and spreads rapidly through communication, commercialization and consumerization.

On the shoulders of giants

Fast IT gets a lot of its value from Core IT. Core IT provides the unique data and product fulfillment that makes Fast IT interesting. Where does Fast IT get the prices for that hot online price comparison campaign? From Core IT. How does fast IT deliver the hot product? By asking Core IT.

An enlightened enterprise will put an API on its Core IT, and wrap it in Fast IT. From the point of view of Fast IT, Core looks like a Web service.