On 27/05/07, Andrew Coppin
Almost all language extensions seem to be of the form "hey, let's see what happens if we randomly change the type checking rules so that *this* is permitted. What would that be like?" Usually it's an extreme struggle to even wrap my brain around what the extension *is*, never mind why this would be a useful thing...
I've read through pretty much all your arguments, and I think they boil down to this: "I don't understand why X is useful, and therefore it can't be useful." Probably the reason why you don't understand why X is useful is because you don't understand X itself in the first place. How can you claim GADTs, existentials etc. aren't useful without understanding what they are in the first place? If you're looking for a good reference to learn these, I suggest the papers apfelmus pointed you towards, or the Wikibook at http://en.wikibooks.org/wiki/Haskell. Moreover, if you've ever written a full-sized Haskell program you'd probably find a use case for at least one of these extensions. For example, you're writing low-level code, and you want in-place array updates. Sounds like a job for the ST monad, which would completely crippled and inherently type-unsafe without rank-2 polymorphism. Or say, as I have done recently, you're writing some forum software, and have a type to represent a Forum, Thread and Post. Now say you want to write some generic code to handle the addition of any one of these -- so this one piece of code allows you to add a new forum, thread or post. Without a splattering of type-system extensions (I used at least MPTCs, FDs and existentials), this isn't going to be possible. -- -David House, dmhouse@gmail.com