editor window force fields

[David M. J. Saslav <saz at OZ>]

In Symbolics 3600 Release 6.0, IP-TCP 29.0, AISite 7.8,
microcode TMC5-MIC 319, FEP 18, on Lisp Machine Apiary-4:

I don't quite know which is the best list to send this to,
and this may be somewhat esoteric even so, but:

when one is in an editor window, and one's mouse is in that editor
window, and it comes to rest anywhere within some relatively large
constant x-displacement (approx.  16 chars when I tried) of the left
wall, it cannot go through that left wall, no matter what, until it
passes out of the region defined by the top of the window the bottom of
the window the left side of the window and the vertical line
x-displacement = 16 chars.  The user must go UP THROUGH THE TOP of the
window or DOWN THROUGH THE BOTTOM of it to get two inches of negative x
displacement, or else "back up" (move to the right) and take a smooth
constant course through the wall.

But except for these, there is no way to get through
what I will now call the force field of the left side
of the zmacs window (which I should of course mention is assumed
to not share its left wall with the left edge of the screen!).

Now I realize why the "this line to top..." mouse feature has some glue
associated to it -- to make it easier to invoke.  From a user's
standpoint this is just right, to an extent.  However, I experienced a case
like the following vertically halved editor scenario:

|~~~~~~~~~~~~~|~~~~~~~~~~|
|             |          |   in which I struggled
|             |          |   with this force field from the right 
|             |Mouse     |   hand window.  It appears to me to be
|             |rests     |   a different issue than
|             |here      |   that of scrolling (top and bottom of
|             |  |       |   windows). (Ironically, the seemingly RIGHT
|             |  |       |   functionality exists with the scrolling--
|             |  V       |   to get through, merely increase the speed 
|             |  x       |   of the mouse).  However, that is not the
|             |          |   issue with the left (or right) side of
|             |          |   a window.  There is a functionality there
|             |          |   that one *might* wish to exploit.  He
|             |          |   shouldn't be *forced* to do something
T~~~~~~~~~~~~~T~~~~~~~~~~T   just because he handled his mouse in an	
|ZMACS        |ZMACS     |   an incorrect manner.
|_____________|__________|

Again, possibly a too-technical point -- I apologize to the people on
both of these mailing lists for taking up so much space of their mail-
boxes (does a mechanism exist for automatic sending to the union of two
mailing lists?) -- but one that I feel has some philosophical signifi-
cance.

dave saslav