Skunks and Unicorns

Out target customers in Insurance aren't generally early adopters of technology, and Insurance-industry buyers are wary of new technologies. A recent Celent study on Legacy and Mainframe Migration noted:

The early 1990s brought the promise of client-server, which was largely ignored by the industry. But the late 1990s brought huge and poorly run projects that were embarked upon using technologies that had not been readied for the insurance industry, such as customer relationship management (CRM) and enterprise resource planning (ERP) projects.

These disasters left a lasting mark on the industry, and trust of technology vendors dropped to a new low.

With this perspective, it's a short-step for Celent to admonish would-be Insurance industry sellers: "don't sell a unicorn", and to conclude that it's a mistake to push leading-edge technologies. All the same, a young tech company (read:Us) looking for explosive growth needs an edge, and the fast pace of technological change makes it imperative to keep an eye on early-adopter technologies, even if most lack immediate applicability for a late-adopter customer base.

To bridge this gap I've been pushing the idea of a "skunkworks" - a playground for us to play with some new ideas and techniques. This blog is a skunkworks effort - we'll see how we might use it, and if we get any benefits from it. Other skunky efforts include:

• The Feature evaluation app, currently at http://triton.clear-technology-skunkworks.com.

• The Future Functionality review, containing a list of Bruce Silver-recommendations for future BPM functionality, at http://jrepko-mob:3003

• The Feature evaluation app, running on JRuby 0.98 (Ruby language in a JVM) at http://jrepko-mob:3004

• A product-proto workflow modeler written in Javascript, at http://jrepko-mob:3006.

I use the skunkworks for experimentation -- not with any expectation that these things will be in the product or our apps anytime soon, but because a number of the techniques used in these playthings (REST, BDD, OpenID, text markup, JRuby, etc. ) might find there way into products, and I don't think it's possible to make a good decision about new technologies without some hands-on time with them.

I have a forum application and an ADM-workflow modeler that should be coming to skunkworks in the near future. I'm open to playing with new ideas and technologies on the skunkworks, and this blog will offer up introductions to new things as we go.