Legacy Code Workshop
We discovered a wide gap between training exercises and real code. We took an enormously long time to prepare a small portion of one out of thousands of modules in our system and get it building in a standalone cpputest environment. Although subsequent efforts are expected to get easier, it is a very steep hill to climb for a very large legacy codebase.
How to improve:
Some offline preparation in identifying and sanitizing modules would have made this workshop more productive. However, the underlying challenges of applying these concepts to a large legacy codebase remains.