Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!snorkelwacker!bloom-beacon!SHAMASH.MCRCIM.MCGILL.EDU!mouse
From: mouse@SHAMASH.MCRCIM.MCGILL.EDU (der Mouse)
Newsgroups: comp.windows.x
Subject: Re: Re: Proposals for Release 5 of X11
Message-ID: <9008150617.AA02464@shamash.McRCIM.McGill.EDU>
Date: 15 Aug 90 06:17:09 GMT
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 40

>>> Proposal: [...] or
>>>           The server should generate an event, when he has redrawn
>>>           the border of a window

>> The former [has a few problems]

>> The latter needs no extensions: simply give the window a
>> border-width of zero and make it a child of another window which
>> sticks out by the desired amount.  Handle expose events on this
>> second window as indications that pieces of the "border" need to be
>> redrawn.

> But, do you really mean to create 2 (or more) windows for every
> (sub-)window within a window hierarchy[?]

No, only for those with complicated borders.  If even that is too much
for you, don't use borders at all - just make the window a little
bigger and draw the object that looks visually like a border as
graphics in the window.

I don't for a moment believe there will be too many windows with
complex borders.  I recall someone quoting 150 as a number of windows
for an application to use.  I shudder to think what this monstrosity
must look like if there are 150 things each of which is heavyweight
enough to require a visually fancy border....

> Even if we accept that a window "is a lightweigth resource", this
> would be too much resource wasting.

Is that a guess or do you have experiments demonstrating it?  I suspect
it wouldn't be as bad as you make it sound.  (No, I have no experiments
to support my guess, but I'm not calling it anything but a guess.)

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu

(For that matter, I never really understood why windows have borders in
the first place.)