[RFC] "http-types" breakage / additions / rework

mikeplus64:

The sentiment of “they knew what they were getting in to, I say, let them crash” is not really accurate: it’s going to cumulatively create a lot of work for people who weren’t the original villains who deigned to not put <X on their http-types dep, but who just want to not have a non-broken package.

If this is so then it’s a strong argument for requiring tight upper bounds on Hackage packages.