>From my previous message: for instance, just using indicate-buffer-boundaries it is impossible to tell whether a buffer is completely empty or whether you are completely overscrolled.
To be more concrete, enable all boundary indicators, including arrows, to the left. Create an empty buffer. The boundary indicators actually give the impression that the buffer is non-empty. Typing RET does not change the indicators, so you can not tell whether the buffer is empty or not. Now overscroll a buffer ending in a newline completely. The "Bottom" indicator now suggests that there is an extra blank line at the end of the buffer, whereas there is not. Note that if you put point on line 4 or line 5 of the scratch buffer (with its default initial content) and do M-0 C-l, the result looks exactly the same. In the first case, the indicated blank line is real, in the second it is fake, but you can not tell the difference. This one is very confusing. Inexperienced users often overscroll completely (by accident), which often confuses them. indicate-empty-lines shows them what happened, whereas indicate-buffer-boundaries gives them the impression that the reason why they were able to scroll this far was the extra blank line at the end of the buffer. I wonder whether it would not be more logical to not show any indicators in an empty buffer and to just show an up arrow in a completely overscrolled buffer. You can not see the end of the buffer, it is scrolled out of view. So why indicate it? Sincerely, Luc. _______________________________________________ Emacs-devel mailing list Emacs-devel@gnu.org http://lists.gnu.org/mailman/listinfo/emacs-devel
