Abstract:
In the second part of the talk, I will present a radically new approach for assessing modular software design. The underlying question is "how can we assess the value of dividing a software system into modules, and why are certain modularizations better than others?" This question is, of course, related to the continuous flow of new solutions and technologies for improving software development, including OOP, AOP, and all generative approaches. While empirical studies can show the benefits of these technologies, they don't provide any insights as for *why* these technologies are beneficial. Recently, I stumbled across an economics-based model of modularity that gave me hope of understanding the hidden issues behind the word 'better.'