Add a bit about various type system concepts

[mark-i-m]

A lot of this is taken straight from Niko's writings here:

• https://gist.github.com/nikomatsakis/fe440aec437f7e8ad4cce8319954a973
• https://rust-lang.zulipchat.com/#narrow/stream/144729-wg-traits/topic/reintegrating.20chalk.20into.20rustc/near/190518417

r? @nikomatsakis

resolves #177 I think

[spastorino]

This looks good!

[mark-i-m]

@nikomatsakis I think I addressed your comments. Could you take a look at the beginning of the early/late bounds chapter again? (basically the last commit)...

[nikomatsakis]

Hmm, the text doesn't seem quite right ..

[nikomatsakis]

I think this is good, should land it :)

[nikomatsakis]

well, I think I would phrase this like so

In Rust, item definitions (like fn) can often have generic parameters, which are always [universally quantified][quant]. That is, if you have a function like

fn foo<T>(x: T) { }

this function is defined "for all T" (not "for some specific T", which would be existentially quantified).

While Rust items can be quantified over types, lifetimes, and constants, the types of values in Rust are only ever quantified over lifetimes. So you can have a type like for<'a> fn(&'a u32), which represents a function pointer that takes a reference with any lifetime, or for<'a> dyn Trait<'a>, which is a dyn trait for a trait implemented for any lifetime; but we have no type like for<T> fn(T), which would be a function that takes a value of any type as a parameter. This is a consequence of monomorphization -- to support a value of type for<T> fn(T), we would need a single function pointer that can be used for a parameter of any type, but in Rust we generate customized code for each parameter type.

One consequence of this asymmetry is a weird split in how we represesent some generic types: early- and late- bound parameters.

[mark-i-m]

I replaced the intro with this (let me know if I didn't replace enough). Sorry, I'm only slowly starting to understand this distinction; you're intro helps a lot.

[mark-i-m]

Oh, huh? I thought I had already merged it! Sorry!