On 2008-08-30, Brandon S. Allbery KF8NH <[EMAIL PROTECTED]> wrote: > On 2008 Aug 30, at 4:22, Aaron Denney wrote: >> On 2008-08-27, Henrik Nilsson <[EMAIL PROTECTED]> wrote: >>> And there are also potential issues with not every legal module name >>> being a legal file name across all possible file systems. >> >> I find this unconvincing. Broken file systems need to be fixed. > > Language people trying to impose constraints on filesystems is the > tail wagging the dog.
I'd say it's just the opposite. The purpose of a filesystem is to hold user data, in ways convenient to the user, which means dictating a usable interface. Dictating the implementation would be closer to tail wagging the dog, though even that's not quite the right metaphor -- it's just a layering violation. The user is in this case GHC or other compiler adopting the suggestion in the Hierarchical modules extension. Just as non-hierarchical file systems have long been considered broken, I think it's safe to now declare that one that doesn't support unicode in some fashion, even if only a userland convention of using UTF-8, is indeed less usable, and hence broken. -- Aaron Denney -><- _______________________________________________ Haskell-Cafe mailing list [email protected] http://www.haskell.org/mailman/listinfo/haskell-cafe
