You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Justifiably, we prioritise the (weakly-)initial-algebra approach to Data.* datatypes via... data definitions, but looking at haskell's Foldable for #2300 I'm conscious that we don't have any implementation of build for List, or more systematically, Church-style analogues for the basic datatypes of the library, together with their build-based equivalence to the standard ones.
This issue might be moot, but it has already come up for me in #1962 as to whether or not to take an initial algebra/free extension approach to Free constructions, or to proceed via a Church-style encoding. I think it's worth further discussion as an issue in its own right, as well as waking up the thread about API vs. implementation in the library design...
The text was updated successfully, but these errors were encountered:
Justifiably, we prioritise the (weakly-)initial-algebra approach to
Data.*
datatypes via...data
definitions, but looking at haskell'sFoldable
for #2300 I'm conscious that we don't have any implementation ofbuild
forList
, or more systematically, Church-style analogues for the basic datatypes of the library, together with theirbuild
-based equivalence to the standard ones.This issue might be moot, but it has already come up for me in #1962 as to whether or not to take an initial algebra/free extension approach to
Free
constructions, or to proceed via a Church-style encoding. I think it's worth further discussion as an issue in its own right, as well as waking up the thread about API vs. implementation in the library design...The text was updated successfully, but these errors were encountered: