Hrvoje, you seem to be Mr. column/line number. Try this: xemacs -vanilla M-x column-number-mode M-x what-cursor-position The above two disagree on whether beginning-of-line is column 0 or 1. Why this inconsistency? An ordinary user would think it's a bug. But I discovered: DEFVAR_BOOL ("column-number-start-at-one", &column_number_start_at_one /* *Non-nil means column display number starts at 1. */ ); column_number_start_at_one = 1; } DEFUN ("current-column", Fcurrent_column, 0, 1, 0, /* Return the horizontal position of point. Beginning of line is column 0. I would vote for just nuking column-number-start-at-one, but why not at least default the value to 0? Then we're consistent with Stallmacs, and more importantly, internally consistent. Martin