Problems with Screensavers and other activity sensitive demons.

[Carsten Haitzler (The Rasterman)]

On Fri, 08 Jun 2007 12:26:22 -0700 Keith Packard <keithp at keithp.com> babbled:

> On Fri, 2007-06-08 at 09:29 +0200, Lubos Lunak wrote:
> 
> >  Ok, I'll try with simpler words :). The XScreensaver extensions is not a 
> > perfectly good one, it's not even a good one. In fact, it sucks, it's
> > broken, and there's apparently nobody to fix it. But I'd indeed like to be
> > proven wrong (and, as I said, KDE after all tries to do the best with it).
> 
> It would be far more helpful if you would propose fixes instead of just
> ranting about the existing state of affairs. Even just a proposed spec
> for a new API would be far better than 'it doesn't work and no-one is
> fixing it'.

indeed. i STILL am wondering what you see is "broken" about it? you can get an
event when screensavers kick in. that is what you want, as a screensaver.
please tell me why you want an idle event instead (and thus each screensaver
program will have to have it sown config as to when it is to kick in, as
opposed to using the x config for that?)

to my mind the problem exists with apps NOT using the xscreensaver
events/extension - inventing new dbus calls, having to exec xscreensaver or a
myriad of other methods to "please disable screensaver for now" is not a good
solution for a system that is primarily run by and controlled by X itself.

xscreensaver has a way to suspend the screensaver cleanly - it is reffed so you
suspend it N times from a client - you neeed to unsuspend N times to go back to
where you were. if a client that asks for suspension is disconnected (dies) -
all "suspends" it had in place go away. nice and clean as well as robust.

> -- 
> keith.packard at intel.com
> 


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