Screensaver to be or not to be (was: Re: Security Audit Request for Screenlocker Branch)

Michael Pyne mpyne at kde.org
Wed Oct 12 21:37:41 BST 2011


On Wednesday, October 12, 2011 20:43:41 Thomas Lübking wrote:
> Am Tue, 11 Oct 2011 21:46:40 -0400 schrieb Michael Pyne <mpyne at kde.org>:
> > Yes. KDE asciiquarium (feel free to look at the copyright headers for
> > that in kdeartwork someday... ;)
> 
> Errr... rather not. The author, *cough* who ever he might be *cough*
> has apparently so far not found the time to implement the resize event,
> so basically it's the only screensaver hack which can *not* simply be
> run in fullscreen mode for sheer entertainment only.

Sure it can. Just click "Test" in the Display & Monitor screen saver options. 
;) There's probably some DBus call to do the same thing.

I will admit that resize is not supported, mostly because scaling down the 
text makes it completely impossible to see. Probably this could work with 
gamma correction but I haven't tried it yet (and no one has ever reported a 
bug about that...)

> If you happen to meet him and he should actually require guidance in how
> to fix that, he may just call back. *cough*

I'm open to suggestions.

> > Yes I do. It's nice to be able to walk away from the computer and let
> > my [...] son move the mouse around or watch the
> > fish move and know that there's at least a half-decent chance he
> > hasn't deleted all my files by the time I got back.
> 
> I think we've a fundamental misunderstanding about "screensavers" here.
> 
> The thread span off because Torgny questioned "no screensaver w/o
> locking" and i replied there's no point in that (since protecting the
> screen isn't required and you can just trigger the show yourself to see
> it w/o any locking facilities)

Sure, but whatever process is waiting to lock the screen normally should be 
the same process to start a screensaver w/o locking the screen. I'm fully 
aware that they can be done independently, and I know that what Martin is 
talking about is just the locking mechanism and not screensavers proper.

I guess my point is pretty much the same actually: Locking and screensaver 
animations should be orthogonal -- it should continue to be possible to kick 
off the screensaver without forcing on the lock.

> You do not walk away and just hope your child won't do any s...tuff
> until the screensaver kicks in and -as a side effect- locks the screen,
> do you? No. You lock the screen (what as a side effect launches the
> undersea show), ensure your son has a good seat and sight and then
> leave, right?

If I walk away then this is exactly what I do. I did used to have the mouse 
corner feature that would enable an unlocked screensaver setup but I disabled 
that due to too many false positives. ;)

> The worst thing that actually *could* happen (i don't think so, but the
> tech details of a final solution are still discussed in the parenting
> thread) is that the locking visuals (w/ compositing, security reasons)
> only work as kwin plugin, ie. the asciiquarium would have to be ported
> once more.

I'm willing to port it if necessary, but like you say the final design is 
still being fleshed out.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20111012/4acb5158/attachment.sig>


More information about the kde-core-devel mailing list