Does CEP improve “Technical Debt”?

Reading Time: 2 minutes

Brenda Michelson blogged on a topic I’ve not come across before – the idea of “technical debt” as a metric of the software development process – and refers to a discussion on this by Steve McConnell. So one wonders: how does a technology like complex event processing impact technical debt a.k.a. software development costs?

Taking Steve’s analysis, we have:

  • time to market (T2M)
  • preservation of capital (PvOC) – i.e. delaying projects as late as possible until revenues better justify the costs
  • delaying expense (DEx) – i.e. retiring older systems later

In CEP one finds that, as a relatively “new” technology area, current software development “best practices” are included in CEP tooling to help with these:

  1. Model-driven-engineering, UML standards, rule and query languages, development aids to minimize T2M
  2. EDA / plugging into existing event buses / sources / networks minimises the startup costs to the benefit of PvOC
  3. Developer tooling and business views / user interfaces for direct business involvement in application updates should make such maintenance of CEP systems more compelling than rewriting, further pushing out DEx.

Of these, point 2 can be expanded as being generic for the concept of CEP (as opposed to the technologies used to deliver it): the ability to declaratively decide what events to process as projects progress, enabling an incremental RAD approach, is probably the most significant at the fundamental level. But I wonder how many project teams have a “Project Cost” counter whirring away in the development office?