If memory serves, what became the BBP was initially introduced at the Haskell Implementors' Workshop in Boston in 2013. (Or, at least, that's when I became aware of it.) I know I'm a special case of a Haskell user, but I've seen mention of the BBP aplenty. However, I do think we should learn from this experience and perhaps establish a very-low-traffic list where we announce breaking language changes.
As for the BBP itself, my 1¢ is that I've never liked it much, but I don't like the status quo either. (Too much `hiding` and `qualified`!) Having neither a good alternative nor time/energy to develop one has kept me relatively quiet in the discussion, and is why I'm offering only 1¢ right now instead of 2¢. (My chief concern is how it affects newcomers to Haskell, which is admittedly a different concern than others have expressed. It may be worth noting that my concern was addressed by suggesting newcomers use the `haskell2010` package... which is now defunct, due to fallout from the AMP.)
As a much more minor point, I believe
> {-# LANGUAGE NoImplicitPrelude #-}
and
> import Prelude ()
should have different behavior. Doesn't the latter import any Prelude instances? Might these be different (and conflicting with!) instances in an alternate prelude? So, I think we must mimic the pragma version as opposed to the empty import list.
Richard
_______________________________________________
Libraries mailing list
Libraries@haskell.org
http://www.haskell.org/mailman/listinfo/libraries