Group colloquium: Defining Software Agility

When: July 19, 2018, 15:45-16:45

Where: HB 2A

Who: Arend Rensink

Software Engineering is a pseudo-science, and that is one of
the reasons why software systems tend to deteriorate over time to a
state of low maintainabilty. Rather than relying on well-understood
phenomena with corresponding measurements, in the analogy of natural
science, in software engineering we have been using metrics that are
thought to be correlated to the quality of software with little (if any)
understanding of the properties themselves that these metrics supposedly
express.

In this (work-in-progress) presentation I will attempt to improve this
situation by inventorising the concepts involved in the notion of
software agility, loosely defined as the "amount of change needed to
adapt an existing system to new requirements". It is a follow-up to the
presentation on "defining software change" that I gave in January. I
hope for your critical attitude and feedback.