When to use MVC and ECB?

Question

According to the book I am reading, I understand this facts. Entity objects is like model objects in MVC which are responsible for maintaining data. Boundary objects are those which interact with external users, handling both input and output. In MVC, user input is detected by control objects, but the handling of output is the responsibility of view objects. That are the difference I have found. But when I do googling, many said that MVC is typically used in user interface design whereas ECB is most often used in business logic. What does that mean? MVC is just used for user interface? If so, what is the responsibility of the Controller and view object in MVC?

Answer 1

The two architectural patterns have a different origin:

• MVC was born from the effort to decouple business logic from user interfaces, in the early days of GUIs. It separates components by purpose, and user-interface was a big one.
• ECB comes from use-case driven development. It separates components based on the use-cases, and these typically capture user/business goals.

The two patterns have some similarities. Indeed, the BCE-Entity components have the same purpose than the MVC-Model components: managing the domain objects.

However there are subtil differences: the MVC-Controller is meant to capture all the user input, and the MVC-View the outputs. The BCE-Boundary is meant to cover the full user interface (input and output) and relate to the BCE-Controler, which coordinates the other BCE components for the purpose of a use-case/goal. In other words, the BCE-Controller is related to the business logic (i.e. the model in the MVC).

If the application covers only one use-case, the main difference will be the controller. But if an application covers several different use-cases, you'll end-up with a different architecture.

In practice, BCE did not really made it through in modern user interfaces, whereas MVC is still very popular thanks to some frameworks.