19
I have a general idea of what existential quantification on types is and where it can be used. However from my experiences so far, there are a lot of caveats that need to be understood in order to use the concept effectively.
Question: Are there any good resources explaining how existential quantification is implemented in GHC? I.e.
- How does unification on existential types work - what is unifiable and what is not?
- In what order subsequent operations on types are performed?
My aim is to better understand error messages that GHC throws at me. Messages usually say something along the lines "this type using forall and this other type don't match", however they don't explain why it is so.