Ben Millwood wrote:
So we don't actually specify the content or API of the library itself, merely state its existence? If we specify the API we make those decisions, if we don't I don't see what exactly you are asking for... So why are you proposing that it be different from a Hackage package in /any/ respect? ...Why doesn't an independent parser do that? ...I don't see any benefit in putting an "official" stamp on a particular one. It doesn't give it a practical edge over any other...
In an enterprise software development environment, getting the software to work in a practical way is not good enough. You need to be able to produce officially recognized documented evidence that it works. The existence of well-specified standards is a prerequisite for the adoption of a language in the mainstream. Haskell is particularly well-suited to the needs of verification, validation, and certification. Galois is doing a lot of great work in this area. I am proposing a simple step, requiring almost no effort on our part. It would immediately raise the quality of the standard considerably, by enriching the expressiveness available to the writers of the standard. It would also immediately raise the perceived value of the standard considerably, by providing the beginnings of an automated validation and certification framework. And it would provide a basis for future work in many directions, both within the standard and based upon it. Thanks, Yitz