Re: TAB from within a text

[Nick Dokos <nicholas.dokos@hp.com>]

François Pinard <pinard@iro.umontreal.ca> wrote:

> Hi, everybody.
> 
> Writing a longish text for my coworkers this morning, I notice that I do
> not know a quick way for collapsing the whole set of paragraphs I'm
> currently writing, when their header happens to be many screenfuls above
> point.  I have to first return to that header and do TAB there.  Even
> this return was not evident to me at first.  I wrongly thought that `C-c
> C-u' would do it, but it jumps far too much and lands one level higher
> than I expected.  Then, /(org)Motion node/ taught me that I could use
> `C-c C-j <up>' to this purpose; which is slightly convoluted to me, as I
> always perceived `C-c C-j' as a kind of sophisticated "reveal".
> 

Checking the navigation menu, I thought that

C-c C-p == outline-previous-visible-heading
C-c C-b == org-backward-same-level
C-c C-u == outline-up-heading

(in particular, C-c C-p in this case) would be the more natural choices for
navigation.

C-c C-j == org-goto does a lot more than that - in particular, it is modal
and you have to get out of the mode, e.g. by selecting a location and pressing
<RET>.

> Is it unreasonable for me to hope that, instead of `C-c C-j <up> TAB', a
> mere TAB from within a long text would quickly do what I wanted?
> 

It does seem unreasonable to me, looking from the outside in: TAB is
overworked, overloaded and much too smart for its own good, and you are
asking it to be smarter and do more :-) But I wouldn't dismiss it as
impossible: org itself came about from a similar idea applied to outline.el
Whether it's worth it, I don't know: C-c C-p TAB has worked for me and has
seemed painless enough to me so as not to go looking for something "better".

Nick