I used to be a huge Textile fanboy. It took quite a bit to drag me (kicking and screaming) to another lightweight markup language. I now use RestructuredText for a variety of reasons. The main reason I stuck with Textile so long is because the output produces such awesome typographic niceties.
In this modern day is it really the job of a lightweight markup language to do ASCII-to- (either) HTML-entity or above-ASCII character substitutions? Certainly, there's quite a historical precedent; Latex has has done typographic-appropriate character substitution for years. Certainly also, it's sometimes nice to have the option for things you know aren't going to have proper character encoding like blog comments (lay-people don't care, and why should they?). But other times maybe this is something that can be easily done just by knowledgeable people writing in UTF-8 and you shouldn't have to rely on a tool to try and do it for you.
I've been thinking a lot about this on-and-off, and would love to see other technical thoughts in the comments. (
(M|N)att?, that's you!) Is anyone out there writing in UTF-8 and putting in the proper punctuation as you go (or using an editor/word-processor that does UTF-8)? Or is it just too much of a pain in the ass? Is it less of a pain if you're not using an American keyboard? (I don't know.)