Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!bloom-beacon!LARRY.MCRCIM.MCGILL.EDU!mouse
From: mouse@LARRY.MCRCIM.MCGILL.EDU
Newsgroups: comp.windows.x
Subject: Re:  Another tiny R5 wish
Message-ID: <9007231140.AA02680@Larry.McRCIM.McGill.EDU>
Date: 23 Jul 90 11:40:43 GMT
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 30

> I'd like to see something which allows Motif to tell the server to
> flash the cursor without having to do it all itself.  Presently Motif
> has to add a timeout handler, receive the timeout event, despatch the
> event, determine the current cursor state, XOR the cursor state, add
> the timeout again, all to flash the cursor.  About two pages of code
> to flash the lights, not to mention the network traffic, the paging
> activity (leaping all over libXM.a, libXt.a, and libX11.a) ...

> Or perhaps this is a protocol change & thus belongs to X.V12?
> Anyway, it amazes me that the server can't flash the cursor.

Part of the problem is that "the cursor" does not exist to the server.
The only cursor the server knows about specially is the pointer
("mouse") cursor.  The text cursor in an xterm window, the
insertion-point caret in a text entry window, they're just a small
graphic doohickeys drawn by the client, as far as the server is
concerned.  So doing this would amount to adding flashing of
more-or-less arbitrary graphics.  Which leads to some way of describing
said graphics...if you're not very careful, you wind up adding an
extension language to the server.  While this is not inherently a bad
thing, it's probably a lot more than you envisioned when you posted :-)

(It's by no means impossible - I'm not out to stifle innovation or
any such thing (though it may sometimes seem like it :-) - but it's not
as simple as "flashing the cursor" makes it sound.)

					der Mouse

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