Re: Draft 1.9c

[raster rasterman com]

On 24 Nov, Paul Warren scribbled:
->  On Wed, 24 Nov 1999 raster@rasterman.com wrote:
->  > On 24 Nov, Paul Warren scribbled:
->  > ->  Hiya,
->  > ->  
->  > ->  I've finally put draft 1.9c up at:
->  > ->  
->  > ->  http://ferret.lmh.ox.ac.uk/~pdw/wm-spec/
->  > ->  
->  > ->  I'd be grateful if people could take a look, and comment.  I have also
->  > ->  stuck up a list of things that still need sorting, linked from that page.
->  > ->  Please let me know if you think of any others.
->  > ->  Currently they include:
->  > 
->  > 1 thing. the _NET_WM_ICON section isnt very explict - infact if i asa
->  > newcomer to the list or the spec jsut read that i'd have no clue what
->  > it means... soit needsto add the fact:
->  > the first 2 CARDINALS are width and height of the icon (respectively).
->  
->  Ooops.  It says "bytes", it should say cardinals..
->  
->  > we dotn allow for multiple icons here - so we coudl xpand the spec to
->  > allow mor eicons to be appended as folows:
->  
->  Erm...
->  
->  _NET_WM_ICON CARDINAL[][2+n]/32
->                       ^^
->  
->  but you reminded of something I meant to say - if there are any parts of
->  the spec which need further explanation or justification in order to make
->  sense to anyone who has not been following this mailing list for the past
->  few months, now's the time to point them out.
->  
->  > as many icons can be provided as an app sees fit - the wm can weed out
->  > the one it wants. it shoudl be suggested to provide the highest-res
->  > (largest) icon first in-case the wm is naievee and only takes the first
->  > icon it finds instead of doing size matching.
->  
->  Yeah.
->  
->  > also some things not in the spec:
->  > _NET_PID
->  > 
->  > there alreayd is an ICCCM machine name hint... but no PID - i really
->  > think a PID hint woudl be infinitely useful so the wm or other apps can
->  > track what process Id that windwo belongs too (VERY handy for nuking
->  > that hung process that doesnt respons to delete events and even if you
->  > xkillclient it the process remains around).
->  > 
->  > this would make taskbars/task managers a whole lot more powerful.
->  
->  Indeed.  The only reservation I have about this is that	it is not
->  network-transparent.  You can only kill processes on the local machine.
->  Although I admit that it would be very useful - get rid of those wretched
->  crashed netscapes...

actually there is alreayd a machine name property as part of icccm:
WM_CLIENT_MACHINE(STRING) = "raster.su.valinux.com"

so u use this in combo with PID (which fills the missing link) and then
the WM or taskbar or whatever can manage all apps that are onthe same
machine - and even coudl spawn helper apps on the particular machine
the app is running on (if it van do this) to try kill or whatevre its
pid there... so as long as we have a machien name AND a PId -we're set

->  Paul
->  
->  

-- 
--------------- Codito, ergo sum - "I code, therefore I am" --------------------
The Rasterman (Carsten Haitzler)    raster@rasterman.com     raster@valinux.com
                                    raster@enlightenment.org raster@linux.com
				    raster@zip.com.au