We should certainly pay attention to this in upcoming usability tests. We still support Alt+Left Arrow, but the objection as I understand it is the accidental invocation of Back, typically when on a form page with the focus on the page. I'm not sure I understand in what scenarios this can be destructive, but I sure agree it can be annoying. We actually used BS in NS 4.x Messenger to go to parent nodes in trees, so we might find data on it in older recorded tests too. I personally thought it was very wierd at the time, accidentally triggering it often, but have easily gotten used to it in the browser, with no ill affects so far. Note that this key is overloaded in IE and Opera as well, so hundreds of millions of people are exposed to it, and may expect it in a browser (right or wrong. Also note there is a large benefit of using it, since it is an easy to hit target, that is already labeled mnemonically.
Peter Brendan Eich wrote: > I don't use Windows, but I agree: I personally believe that Backspace > is the wrong key to map to Back -- it violates all UI rules of thumb I > know (e.g., don't overload destructive gestures with other meanings, > whether destructive in a different way [as Back can be]). I recommend > you petition in mozilla.ui, cc'ing [EMAIL PROTECTED] at least, for > some usability data to justify this change. There very likely have > not been any usability studies, and lacking real data, I think Mozilla > should do what Netscape 4.x did in this case -- IIRC, 4.x on Windows > used Alt-< as the shortcut for Back.
