"In general relativity, an event horizon is a boundary in space-time beyond which events cannot affect an outside observer. In layman's terms, it is defined as "the point of no return", i.e., the point at which the gravitational pull becomes so great as to make escape impossible, even for light. An event horizon is most commonly associated with black holes."

In a previous post I attempted to make the case that not enough information makes it out of the development process turning a normally black box software process into a black hole. This is problematic for us and the both the upstream and downstream processes as detailed information can provide guidance when making strategic decisions. By producing a black hole when developing software we actually limit our ability to positively influence the outcomes of work we own.

The question then is can we send signals upstream, beyond our Event Horizon as it were, to decision makers via a series of metrics, that will allow them to make better decisions?

Several of my colleagues, including my good friend Larry Oesterreich, are engaged in creating and disseminating these metric. Larry in particular has recognized some interesting code smells that need to be cleaned up, and of course he is right. The thing I think is being missed is that these signals alone do not guarantee the right choice will ever be made, it will only make sure all concerns are correctly communicated. Isolated metrics mean little without an accompanying heterogeneous analysis.

This is the part where our development team has to move outside of its comfort zone and understand the motivation for what appear to be arbitrarily bad decisions earlier in our development pipeline. So while you spend time understanding your own metrics you also have to understand what motivates the Support Manager, and the Sales and Account Executives. Their metrics may look nothing like ours (the Event Horizon is complex) but not being able to compare these concerns to entrenched developer concerns is simply a failure of imagination.