> My personal view is that denotify should be yanked from the specification. > I > see no more reason for it than I see for having defileinto or deredirect. > (The > last one has an especially nice name, doesn't it?)
I'd be happy to see it dropped too, I doubt very much that I'd ever implement it even if it doesn't get dropped.
It appears to me to be fairly easy to implement so I'd probably do it in order to claim conformance, but I cannot imagine a case in our environment where it would actually be useful and used. And unused (or rarely used) options are often a place for bugs to congregate.