2014 Applicative => Monad proposal

[David Luposchainsky]

On 2013-05-24 23:59, wren ng thornton wrote:
> FWIW, as things stand, I'm also -1 for Pointed. Pointed doesn't really
> have much meaning on its own, so I think it's more helpful to invert the
> hierarchy as Kmett has done and have Apply/Bind classes as the
> generalization of Applicative/Monad.
> 
> More to the point, the Pointed proposal doesn't meet the AMP goals of
> gentility and compatibility. And the controversy renders it dubious for
> inclusion in the proposal to make a unified compatibility-breaking change
> to clean up the long-standing complaints.


I'd also vote -1 for Pointed. For one Applicative and Monad are widely
used, can't say that of Pointed. Compatibility is another big thing: the
AMP connects things that are already there, while Pointed would mean
introducing the typeclass in Base in the first place.

On the other hand, I think the AMP can also serve as a model for similar
proposals in the future. Should it turn out in five years that Pointed
is amazing and we all need it, its inclusion won't be much different
from the AMP right now, except that we'll all know what to expect
beforehand.

David