Luka Jacobowitz
As for other things to read, the only thing I've got is the inspiration from PureScript: https://pursuit.purescript.org/packages/purescript-validation/3.0.0/docs/Data.Validation.Semiring
This is really cool, thanks for sharing!
/cc @stephen-lazaro
I think this will have to wait until 2.0, so no need to hurry, we've got tons of time :)
I fixed the hierarchy, not sure what I was thinking there :smile:
To keep breakage as small as possible it might be good to either 1. Wait with the `Decidable` PR until 2.0 Or 2. Add the invariant super classes now and...
@catostrophe Everything you're saying makes absolute sense to me, thanks a ton! Though I think we should discuss the issue of renaming existing type classes in a different issue, otherwise...
Of course with `Nothing` come some very annoying problems, type inference regularly breaks when the signature contains `Nothing` somewhere and of course we run into the *dead code* issue as...
I added a WIP for the missing type classes (sans `Decidable`) here: https://github.com/typelevel/cats/pull/2632
@catostrophe do you mean `point`? I'm not sure I can deduce what's wrong with it. Also I fully agree with your assessment on the two possible `InvariantMonoidal` instances for `Predicate`....