[GHC] #1968: data family + GADT: not implemented yet

[GHC]

#1968: data family + GADT: not implemented yet
------------------------------------------------+---------------------------
 Reporter:  Remi                                |          Owner:  chak            
     Type:  bug                                 |         Status:  new             
 Priority:  normal                              |      Milestone:  6.10 branch     
Component:  Compiler (Type checker)             |        Version:  6.9             
 Severity:  normal                              |     Resolution:                  
 Keywords:  data type family GADT choose_univs  |     Difficulty:  Moderate (1 day)
 Testcase:                                      |   Architecture:  Multiple        
       Os:  Multiple                            |  
------------------------------------------------+---------------------------
Comment (by chak):

 Yes, as Simon says, you are perfectly right about the three examples.

 Just as a final remark on that topic, the reason why it is tempting to
 think that "naming" a parameter in a `data family` declaration makes a
 difference is because it '''does''' make a difference for a `type family`.
 This is a point where data families and type synonym families are
 fundamentally different.  The underlying type theoretic rational derives
 from the fact that a data family always introduces a '''new''' type,
 whereas a synonym family doesn't.  A secondary consequence is that data
 families may be applied partially, whereas a synonym family must always be
 applied to at least as many type arguments as its arity (i.e., the number
 of named parameters in the declaration).

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1968#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler