> My preference would be that while in manual "scrolling mode" that output
> not cause the screen to scroll unless [...] the currently displayed
> portion of the buffer is being rotated/forced out of existance by the
> new output.  [...] this feature is a necessary evil if we want to
> avoid having systems be frozen in place by being left in scrolling
> mode.

Not quite.

It would also be possible to _never_ disturb what's being displayed.
In the circumstance you describe, there would then be a ringbuffer of
output scrolling, plus one screenful frozen for display.

Of course, this then calls for some interface design thought to figure
out what to do when you scroll after having reached this state.

The major benefit of this, to me, is that you can scroll back and leave
something of interest displayed, and it will never get scrolled into
nonexistence.  Whether this is something enough people care about to
bother with, never mind being worth the additional code complexity it
would call for, is of course another issue.

I also think it would be good to distinguish between "not scrolled" and
"scrolled back by zero lines".  Both display the same screen contents,
but the former scrolls when a newline is printed, whereas the latter
doesn't (any more than if you'd scrolled back it would scroll on
output).

But this is all armchair quarterbacking. :-)

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B