On Thu, 09 Jan 2014, David Holland wrote:
> Can ghc be built using some other Haskell implementation? Not AFAIK. But even if it theoretically can, the others (hugs and nhc98) don't really work well enough to do this in practice.
They don't have to work well enough for normal use. If otherhaskell compiles ghc very slowly, or only on some platforms, or if ghc-compiled-by-otherhaskell runs very slowly, it can nevertheless be good enough for occasional use in verifying the results using the technique described by David A. Wheeler at <http://www.dwheeler.com/trusting-trust/>
--apb (Alan Barrett)