Continuous testing
In an ideal world, continuous integration will enable us to adopt continuous testing, which provides us with constant and early feedback. This is essential because our bounded contexts and the resulting domain models are in a constant state of evolution. Without the bedrock of a stable suite of tests, it can become very hard to sustain a reliable process. Approaches such as the test pyramid, testing trophy, honeycomb, and so on are acknowledged as reasonable ways to implement a sound continuous testing strategy. All of these approaches are based on the premise that a large number of cheap (computationally and cognitively) unit tests form the foundation of the strategy, with the number of tests in other categories (service, UI, manual, and so on) reducing as we move through the chain.
However, we are in this new world of fine-grained components that work by communicating with each other. Hence, there is a bigger need to verify interactions at the periphery in a...
