MMCM@MIT-AI 02/07/80 21:53:48 Re: glossary
complementing blinker is really redundant, all blinkers are by nature complementing
in that they must xor to be undoable.
 
RWG@MIT-MC 02/05/80 21:53:04
To: ACW at MIT-MC
i think i found it: WINDOW MANUAL .  typo: descandant.
   
ACW@MIT-AI 02/04/80 04:24:43 Re: Window hierarchy arcana.
To: HIC at MIT-AI
CC: MMCM at MIT-AI, MOON at MIT-AI, ACW at MIT-AI
I found the following weird code fragment.  Needless to say
it is a part of WINDOW-PUSH, which I have been trying to grok
for the last day or so.  This code fragment indicates that I
don't know everything I thought I knew.

(SETQ ACTIVE-P (MEMQ WINDOW (SHEET-INFERIORS (SHEET-SUPERIOR WINDOW))))

How can a window not be an inferior of its own superior?

   Yours,
     ---Wechsler

Date: 4 February 1980 06:47-EST
From: Howard I. Cannon <HIC at MIT-MC>
Subject:  Window hierarchy arcana.
To: ACW at MIT-AI
cc: MMCM at MIT-AI, MOON at MIT-AI

     Keep in mind when reading WINDOW-PUSH that it is really a kludge.

In any case, the question you raise is an important one.  You have
undoubtedly come across "activation".  Well, a window is said to
be active if it is on it's superiors list of inferiors.  This basically
means that the window is one which the system is interested in.
A deactivated window, on the other hand, does not appear on its superior's
list of inferiors.  This means that it is invisible to the system,
and if noone has any pointers to it it will be GC'ed.  Deactive windows
cannot be exposed (though this is hidden by the fact that there is an
expse daemon that conspires to activate the window when the expose message
is sent).

Does this clear things up?

  
RWG@MIT-MC 02/04/80 23:27:15
To: ACW at MIT-MC
rumor has it that you have an elegant thumbnail description of
flavors in a document you are constructing.  pointer?
 
Date: 1 February 1980 14:08-EST
From: Howard I. Cannon <HIC at MIT-MC>
Subject:  WINDOW-PUSH
To: ACW at MIT-AI

The intention in window-push was presumably to copy the rest arg from tXe
stack so it could be get-propped/putprop-ed.  The correct code is
probably
%AUX INIT-PLIST)
(setq init-? (copylist ini-?))
(setq init-plist (locf init-?))  ; I DOn't remmeber the name of the rest arg

The property list stuff treats locatives as disembodied property lists
(not lcoatives, really, but the list they point to).


HIC@MIT-AI 01/31/80 02:15:34
"No window is exposed it its superior is not exposed" is wantonly
false.  A window doesn't really know its absolute lcoation in memory -- rather,
it has an array that it can write into to change bits.

The latter stuff in WINDOW MANUAL is really on the right track.  Too bad
IAP is over...
    
Date: 13 January 1980 19:54-EST
From: Mike McMahon <MMCM at MIT-AI>
Subject:  New window system documentation
To: MOON at MIT-AI
cc: HIC at MIT-AI, ACW at MIT-AI

    MOON@MIT-AI 01/13/80 01:30:34 Re: New window system documentation
    Please read and check the file AI:LMWIN;OPTION INFO.
I typed in some more explanations.
  
Moon@MIT-AI 01/15/80 16:26:25
To: ACW at MIT-AI, HIC at MIT-AI, MMCM at MIT-AI
The function SI:MESSAGE-REPORT in the file AI:MOON;MSGRPT makes a listing of
all the messages in a given system of flavors, and what their arguments are.
It may prove useful as a starting point for documentation.

Moon@MIT-AI 01/17/80 09:50:28 Re: New window system
I forget if I told you about the file LMWIN; SHEET MSGS, an automatically-generated
list of all the messages and the arguments (argument names from the primary method
for the most basic flavor handling the message.)

MOON@MIT-MC 01/26/80 11:05:36 Re: New window system
To: HIC at MIT-MC, MMCM at MIT-MC, ACW at MIT-MC, MOON at MIT-MC
It just came to my attention that you changed the arguments to the
:set-cursorpos stream operation.  I guess the change is reasonable;
this note is to make sure that it remembers to get prominently documented.