poscat0x04 · January 10, 2020 14:02 · poscat0x04 · Jan 10, 2020
diff --git a/Why functional matters b/Why functional matters
 This is my philosophical argument on why functional programming matters.

 1. Abstraction
 Functional programming unlocks a whole varieties of abstractions (monads, arrows and profunctors for example) that are meaningless to
 mimic or impractical and unable to implement in imperative languages. These abstractions help us to tackle and understand a complex 
 software system to effectively reduce the complexity
 (insert notation as a tool of thought citation)

 2. Constraints
 Funcitonal programming let us specify what exactly a program can do and cannot do -- type signatures, class constraints, dependent types,
 algebaric data types, seperation of pure and impure, refinment types... These can help us (with the aid of a compiler) to eliminate 
 as many errors as possible at compile time.

 3. Modularity and Composability
 Functional programs are highly composable and modular and thus reusable. [why fp matters] Some examples are parser comibnators and
 software transactional memories

 4. Theoretically simple
 Functional programming languages can be formalized easily with type theory.

 These categories are not orthogonal to each other and there are some overlaps (parametric polymorphism for example can fit into both 2 and 3)
	This is my philosophical argument on why functional programming matters.

	1. Abstraction
	Functional programming unlocks a whole varieties of abstractions (monads, arrows and profunctors for example) that are meaningless to
	mimic or impractical and unable to implement in imperative languages. These abstractions help us to tackle and understand a complex
	software system to effectively reduce the complexity
	(insert notation as a tool of thought citation)

	2. Constraints
	Funcitonal programming let us specify what exactly a program can do and cannot do -- type signatures, class constraints, dependent types,
	algebaric data types, seperation of pure and impure, refinment types... These can help us (with the aid of a compiler) to eliminate
	as many errors as possible at compile time.

	3. Modularity and Composability
	Functional programs are highly composable and modular and thus reusable. [why fp matters] Some examples are parser comibnators and
	software transactional memories

	4. Theoretically simple
	Functional programming languages can be formalized easily with type theory.

	These categories are not orthogonal to each other and there are some overlaps (parametric polymorphism for example can fit into both 2 and 3)