Power-off stupidity remains in Fedora 17

[Sam Varshavchik]

Olav Vitters writes:

> On Fri, Jun 15, 2012 at 08:27:22AM -0400, Sam Varshavchik wrote:
> > Olav Vitters writes:
> >
> > >You're very demanding and call people "royal highnesses". Sure seems
> > >like you seem obliged that people do what you think is right (other
> > >idea? wrong!) on a timeframe that you think is correct.
> > >
> > >Anyway, there I go again, being a royal highness :P
> >
> > Well, you've pretty much ignored everything else I wrote. Including,
> > specifically, the reason why the shoe fits, here. Really, can't see
> > a better way to prove my point.
>
> Of course I am here to fix your every problem. And of course I am not
> doing the adult thing.
>
> Just kidding. Yes, GNOME shell using 40% CPU and so on is not nice and

This was only mentioned briefly in the last paragraph of my message, most of  
which was about the monitor DPI resolution breakage.

> should be fixed. However, your argumentation is weak. There are loads of
> ideas and loads of bugs that haven't been fixed. Turning that into proof
> that GNOME/developers/etc must be "XXX"... cannot take such
> argumentation seriously...
>
> Saying things like "adult thing to do"... blergh.

That was not said about the gnome-shell eating 40% CPU, by the way.

But, I'm afraid I simply don't know how else to describe it, when it's now  
two years and counting, since a setting was needed in custom.conf that gdm  
itself doesn't really need to do anything with, except to pass it along as  
an additional option to Xorg. What even makes it more depressing, IIRC, the  
setting was there in the first place, for many years, before it was thrown  
away to prevent all the stupid and dumb users from doing The Wrong Thing,  
since gdm now knows better than them:

root      1535  1517  0 Jun09 ?        00:00:00 /usr/libexec/gdm-simple- 
slave --display-id /org/gnome/DisplayManager/Display1
root      1541  1535  1 Jun09 tty1     02:17:35 /usr/bin/Xorg :0 -background  
none -logverbose 7 -auth /var/run/gdm/auth-for-gdm-weJJt3/database -seat  
seat0 -nolisten tcp vt1

If that's the options that Xorg needs to get started with, by gdm, then they  
just have to be right, don't you know? If someone needs to have -dpi 140  
added to them, and the user is willing to manually edit a conf file, and put  
it there – well then, the user must be WRONG, no doubt about it, and gdm  
must prevent the user from making a horrible mistake like that, until this  
can be fixed the proper way, perhaps in another three years.

At least that was my interpretation of the few sentences worth of a reply  
from Gnome, on this issue.

Understood. It took me a while, but I was finally enlightened. I finally  
figured out why it's not really a bug – it's because NoSquint can be  
easily installed in Firefox. Looks like most folks affected by this have  
figured out the workaround for Firefox, given the stats on mozilla.org; and  
all other apps that need to run under Gnome can simply implement the same  
workaround, by themselves. Having every appp implement a version of NoSquint  
makes so much more sense than gdm passing through the right option to Xorg!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.fedoraproject.org/pipermail/users/attachments/20120615/b9bc014d/attachment.sig>