15 minutes ago, Guillaume Marceau wrote: > On Thu, Jul 14, 2011 at 1:42 PM, Eli Barzilay <e...@barzilay.org> wrote: > > Yes -- paredit is exactly an attempt to get the always-balanced > > benefits of structured editing, without an actual structure editor. > > Yes, but paredit still gets its priorities backward. You want to > have the most commonly used edit operations on the easiest chords > (or without any chord at all, if you are trying to avoid RSI.) In > paredit, keystrokes to edit sexp are heavily chorded, and keystrokes > to edit single-characters are simple, even though single-char only > occur when fixing typos in identifiers, which is quite rare when > using auto-completion.
I'm sorry, but this is a completely nonsensical complaint. paredit does not have any priorities of its own here, it is simply using the same exact key (-chords) that are used by default on Emacs. I guess that you can say that it has a priority: follow the standard Emacs keys. Given that it *is* an Emacs extension, doing anything other than this would be insane. Case-in-point: > Case-in-point, > Moving forward by one sexp in paredit is CTRL-ALT-F (ouch). In > DivaScheme it's just L. > Moving forward by two chars in paredit is RIGHT, RIGHT. In DivaScheme > it's W, RIGHT, RIGHT if paredit would have used vi-style modes, it would have been dead and burried years ago. In fact, *you* said that DivaScheme "wasn't very popular with the healthy-wristed". I appreciate the potential pain etc that people with RSI go through, but this is completely unrelated to what paredit does. Keeping it separate is important -- this way, you need one solution for structure editing and another for dealing with RSI -- and the only requirement on both sides is that they compose well. (In addition, RSI is not the only problem that affects keyboard use. What about different keyboard layouts? Different input devices?) -- ((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay: http://barzilay.org/ Maze is Life! _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/dev