soul-searching what we see after C-x 2 M-> C-x 1

[Dan Jacobson]

>>>>> "Jason" == Jason Rumney <jasonr@altavista.net> writes:

Jason> jidanni@deadspam.com (Dan Jacobson) writes:
>> Where's the crushing admission that there's a bug?  I see none posted.
Dan> $ emacs -q .

Jason> Note where the cursor is.

Dan> C-x 3

Jason> Note that Emacs goes to great lengths to keep the cursor location
Jason> visible in both windows.

Dan> C-x 1

Jason> Note that Emacs goes to great lengths not to change your view of the
Jason> buffer unnecessarily.

Dan> The stuff is all moved over to the side, no? 

Jason> No, it stays EXACTLY where it was before you pressed C-x 1

>> Even though my brain is Limited Edition, I'm sure I found a bug.

Jason> I'm sure you would have found a bug there whatever behaviour it 
exhibited.

Damn right, as mother will not be pleased unless my bug quota is
reached.  Consider: it is an average day, you are editing an average
file, you split the window into two with a C-x 2.  OK, now you place
the cursor into the lower window, do M-< to go to the top of the file,
and then do C-x 1.  [unfair example maybe, but] does emacs hesitate to
make the most out of your window space---doing the "right thing", no?

[oops, instead when doing M-> in the top window it acts like you
say]... anyways, I say that C-x 1 and similar [C-x 0] commands need "a
whole new awareness of the user's priorities as to how the position of
the buffers in the remaining windows should look".  For further
instance: why is it smart to show me the never-never land beyond the
end of a file upon C-x 2 <escape> > C-x 1
occupying 3/4 of the screen, instead of say, file: top 3/4 of screen,
never-never land: last 1/4 of screen?

Yes, the cursor should stay at the bottom of the file, but the file
should still occupy at least half the screen please....

Do this simple test on a file longer than your window:
C-x 1 M-< M-> C-x 2 C-x 1

Notice that just before we did C-x 2, the file was occupying oh, 85%
of the screen.  However, when we finish, it only occupies oh, 20% of
the screen.  How am I to consider Emacs as my best friend with
behaviour like that.  Explain why it is so warmly aware of what I
want to see upon the M->, and then callous upon the final C-x 1?
-- 
http://www.geocities.com/jidanni/ Taiwan(04)25854780