Re: Updating plans of lang/ghc

[Marc Espie <espie%nerim.net@localhost>]

On Fri, Jan 10, 2014 at 04:02:26AM +0000, David Holland wrote:
> On Thu, Jan 09, 2014 at 09:42:14PM +0200, Alan Barrett 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.

> No. But... compiling ghc is probably the most demanding thing any
> Haskell implementation is likely to be asked to do. So they actually
> have to work better than for normal use. :-/

"Work better", nice euphemism.

Let's call a cat a cat. Haskell is a lab experiment gone wrong.

There's *no* production quality haskell compiler. Heck, even ghc has
an atrocious bootstrap process. And have you looked at its signal
handling ? It tends NOT to like ^Z/continue. I've hung hundreds of
compilers that way.