Explain different types of cohesion and coupling in system design.

Cohesion and coupling are two key characteristics that are used to evaluate the quality of a system design. They are often used in the context of object-oriented programming and software engineering.

1. Cohesion: Cohesion refers to the degree to which the elements of a module belong together. In other words, it measures the single-mindedness of a module. There are seven types of cohesion, listed from worst to best:

   • Coincidental cohesion: It's when parts of a module are grouped arbitrarily.
   • Logical cohesion: It's when parts of a module are grouped because they are logically categorized to do the same thing.
   • Temporal cohesion: It's when parts of a module are grouped by when they are processed - the time they are processed.
   • Procedural cohesion: It's when parts of a module are grouped because they always follow a certain sequence of execution.
   • Communicational cohesion: It's when parts of a module are grouped because they operate on the same data (i.e., they communicate with the same data).
   • Sequential cohesion: It's when parts of a module are grouped such that the output from one part is the input to another part like a pipeline.
   • Functional cohesion: It's when parts of a module are grouped because they all contribute to a single well-defined task.

2. Coupling: Coupling is the degree to which one module depends on the other modules of a program. It is a measure of the strength of the interconnections between modules. There are five levels of coupling, listed from best to worst:

   • No coupling: Modules do not communicate at all.
   • Data coupling: Modules communicate using only data parameters.
   • Stamp coupling: Modules communicate using composite data structures.
   • Control coupling: One module controlling the flow of another, by passing it information on what to do.
   • Content coupling: When one module uses the code of another module, such as a branch.

In general, a good system design will aim for high cohesion (preferably functional cohesion) and low coupling (preferably no coupling or data coupling).