inspired by Ted Neward’s original Architectural Katas
Fitness Function katas are intended as a small-group (3-5 people) exercise, usually as part of a larger group (4-10 groups are ideal), each of whom is doing a different kata. A Moderator keeps track of time, assigns Katas (or allows this website to choose one randomly), and acts as the facilitator for the exercise.
Evolutionary architecture allows different parts of the system to evolve in the ways most sensible to solve a problem. However, we don’t want the system to evolve in a way that harms some architectural concern. For example, improving performance with caching might accidentally harm security in the process. How can we build guidelines within the architecture to support change but guard specific attributes? Evolutionary computing defines a fitness function as an objectively quantifiable function used to summarise how close a given design solution is to achieving the set aims. When defining an evolutionary algorithm, the designer seeks a “better” algorithm; the fitness function defines what “better” means in this context.
As defined in our book Building Evolutionary Architectures, we borrow this concept to for architectural fitness functions:
An architectural fitness function provides an objective integrity assessment of some architectural characteristic(s).
Fitness functions encompass existing engineering practices such as testing and metrics, but also advanced capabilities such as Netflix’s Chaos Monkey A wide variety of dimensions exist for architectural fitness functions: context, cadence, scope, and so on. We identify a variety of important fitness function dimensions and how they interact, along with techniques for discovering them.
Doing an Fitness Function kata requires you to obey a few rules in order to get the maximum out of the activity. However, one rule trumps all the others: Any other questions that are not already covered by these rules, you may ask the Moderator about. When in doubt, ask.
The first step is to assemble your project team
Once that’s done, select your kata and proceed on to the Discussion Phase.
For the next “N” minutes or hours (depending on what the Moderator tells you), your project team will now examine the requirements for the kata as given, and work out a rough vision of how to solve this architectural problem. We’re not expecting code, but rather a strategy for how you would attack this particular verification of this architectural element.
The Moderator will give you some kind of audio and visual cue when your time is getting low, and then at some point, time will run out and it will be time to move on to the Peer Review Phase.