Re: [Haskell-cafe] Where are rank-3 types necessary in practice for maintaining abstraction?

Am 03.04.2018 um 08:54 schrieb Timotej Tomandl:

Ok, I thought about it a bit more and realized mask in Control.Exception is the one where rank-3 is necessary, which is the example I was looking for.

Given your newly acquired insights, do you expect that there will be ultimately a valid example for higher ranks? Is there a theoretical limit? Or a practical one? E.g. it might be too awkward to mentally handle higher-rank polymorphism - or maybe there's no real mental difference when dealing with higher-ranked polymorphism, I don't know because I didn't have to deal with that yet, so I'm curious.

Sorry for the spam.

Actually it was interesting. Regards, Jo