Possible bug in ghc 4.08.1, I leave it to you

[Simon Peyton-Jones]

| > But where is this requirement in the Haskell report? It 
| might be that I
| > don't understand what the report says, but this seems like 
| a perfectly
| > good use to me.
| 
| See 4.5.2 of the report and section 11.6.3 of Mark Jones' 
| "Typing Haskell
| in Haskell" (http://www.cse.ogi.edu/~mpj/thih/ ) paper for a 
| discussion of
| the issue. The conclusion there is that the H98 restriction 
| on the shape of
| signature contexts in mut-rec groups is overly restrictive. 
| Last I heard
| (about 12 months ago) was that Simon PJ _might_ consider treating this
| as a bug in the H98 report & change it.

I did think about this some while ago, and I'm very disinclined to fiddle
with
the H98 Report on this one.  The Report is perhaps too conservative,
but that's a good side to err on.

Still, I am interested to see a Real Example where it makes a 
difference.  The only examples I've seen are artificial ones
like the one you give.  

Simon