This email is in response to the (ongoing) Haskell-cafe discussion about type declaration syntax. There are a number of hypothetical changes to Haskell that could be aesthetically pleasing or convenient if implemented, but not to such a degree that the added convenience outweighs the added complexity of a pragma-gated feature or the cost of breaking existing code. (Case in point: renaming type/newtype/data to be less confusing to first-time users) However, there may be better opportunities to implement such changes in the future, and it would be a shame if every good (but perhaps somewhat trivial) idea was lost to the annals of the mailing list. Therefore, I proposed that we might maintain a list of nitpicks, which could be used to inform future decisions about breaking changes to the Haskell specification. I'm not sure what the most appropriate forum is for such a list. The mailing list is sub-optimal, because we might end up with a huge email chain littered with bikeshedding reply-alls. I would be willing to host and maintain such a list on Github, but that seems somewhat antithetical to the Haskell community's generally more distributed style of discussion. The wiki might be appropriate, but I don't know if it's conducive to proper moderation; we may have to filter out suggestions like "add for loops". --Will