Re: Still need a hint for undecorated windows

[Carsten Haitzler (The Rasterman) <raster rasterman com>]

On Thu, 23 Jun 2005 18:36:22 +0200 Philipp Lohmann - Sun Germany - ham02 -
Hamburg <Philipp Lohmann Sun COM> babbled:

> Let's not make this into a flamewar. I guess i will not convince you
> anyway. But, please, have a little place in your heart for us
> application developers that occasionally actually need the window
> manager to do what they need. In the end what good is a window manager
> without applications ?

don't worry - i'd say a lot of wm devs do appreciate app devs and their needs.
:) sometimes we just need to sit in a room and discuss it and come to an
agreement. hell - sometimes us wm devs wish app devs would not be so evil and
cause us to do insane amounts of "work around code" :) sometimes it'd be nice
for app devs and wm devs to have a gentleman's agreement to just drop parts of
icccm because they suck so hard! :P but i think we do all need to work together.
borderless windows are trivial to implement. i dont see any argument from the wm
author side saying "dont ask us to do this - its really painful to do". it's
just a semantic ui design/consistency etc. issue and that is another matter
entirely :) for that a wm has the ability to allow a user to override the apps
wishes and say "app X - you will sit in THIS corner at THIS size and that THIS
border - like it or not!" that's the wm's job to provide the ability. :)

personally i wish there were better ways to uniquely identify a particular app
consistently - netwm's pid property, name, class, title and maybe netwm role
combined make for quite a mess. i was thinking that in future maybe we can have
some registration system where app devs get a unique ID for their app and can
then use that. an md5sum of the binary would be nice but an upgrade would break
things - or a recompile. but for now let's assume md5 sum of the binary - and we
place this as a property on a window... now we can be guaranteed to UNIQUELY
identify THAT app. but we need something better than an md5 of the binary for
the ID. it'd also be nice if this can be slid into xlib itself so that legacy
apps suddenly "just work". any one got ideas?

> Kind regards, pl
> 
> -- 
> The major difference between a thing that might go wrong and a thing
> that cannot possibly go wrong is that when a thing that cannot possibly
> go wrong goes wrong, it usually turns out to be impossible to get at
> or repair.   -- Douglas Adams
> 
> _______________________________________________
> wm-spec-list mailing list
> wm-spec-list gnome org
> http://mail.gnome.org/mailman/listinfo/wm-spec-list
> 


-- 
------------- Codito, ergo sum - "I code, therefore I am" --------------
The Rasterman (Carsten Haitzler)    raster rasterman com
裸好多                              raster deephackmode org
Tokyo, Japan (東京 日本)