Less types and more functions.

quote.txt

As lisp programmers have know for a long time it is better to have a smallish number of ubiquitous data types and a large number of small functions that work on them, than to have a large number of data types and a small number of functions that work on them.

I dont think I am sure what that means. From what I guess is that its better to have a file und represented as string and have functions like the dotnet file api instead of a type like FileInfo. (Yes i know FileInfo id very OOish)

It doesn't say you should use string. It's better to have a separate file data type that can be distinguished from string. But yes we want to separate the data from the functions.

Well functional programming in general follows the principle of building larger things from small basic, reusable units.

If you take a look at the datatypes used in FP, they don't follow a hierarchical model (like inheritance) but are rather composed out of the core basic datatypes (List/Array/Tuple/Hash). The heart of every FP language is a core group of function definitions that can work on those data types .... and there are a lot of them :-)

Even the Record abstraction in Elixir for example is in the end just a Tuple and can be used with every function that can handle a tuple.

It's better to have a separate file data type that can be distinguished from string.

But of course all string functions should also work on the "Content" property of the file. But a file is usally more than just a string. You have a FileName and maybe other functions which allow to process the file sequentially. You don't want to load the big files completely into the RAM.

But in order to understand the statement keep in mind Lisp has no strong typing.

All Lisp functions should work on the generic data types like array, map,...

If you look at ML, F# or haskell then you'll see that you can have both. You can define generic functions which work on complicated data structures and keep the types distinguishable.