Do you know whether this is because GHC and Hugs are merging? For some reason I have the impression that they are getting closer, but I can't remember what I read.|
Looking at that first list again, now that I know a little more, I wonder if point 3 reflects the apparent difficulty in parsing Haskell. Again, I cannot remember the source, but I believe Haskell's syntax, although much clearer than ML's to a human, is much harder to parse (I guess all those annoying little keywords like "in" in ML make parsing easier). That would make ML parsers more robust, allowing more error messages.