That’s it! We have learned all about many different concepts in a tour de force, which we would like to summarise once again. Our starting point was a problem. Whenever we manage to understand the solution to a problem as the transformation of input data into output data, software is a good candidate for the solution. The relationship between inputs and outputs can be described as a (mathematical) function. That’s the what of the problem solution. Algorithms are prescribe how to compute such functions. An algorithm can then be implemented by different programmes. These run on hardware, in particular a processor. Alternatively, individual functions can be implemented using machine learning. Software typically consists of multiple programmes communicating with each other. The central activities of software engineering are: Identifying and understanding the right problem, breaking it down into sub-problems, establishing a structure for solutions to these sub-problems, ensuring and verifying correct functionality and quality, and evolving the software system over time. Finally, software is everywhere, in the cyber-physical systems that surround us.
I think we should strive to understand, at least in principle, what this is all about. Hopefully our text can contribute to that understanding.
Peter Bludau and Johannes Kroß from fortiss; Manfred Broy, who would have preferred to see a calculator instead of a cake; Severin Kacianka and Traudl Fichtl from TUM; Niina Zuber and Jan Gogoll from bidt; and Alexandra Pretschner from Boosting Change helped a lot to simplify my descriptions of the complex material. The idea to apply the cake baking example to the automated bakery is from Gordon Mühl from Huawei. Many thanks!