One main difference between x11 and wayland is that in x11 all applications receive all input whereas in wayland the compositor receives input and decides which application may receive it.
This causes a major issue with application which need input grab for running or showing a nested desktop. This affects e.g. GUI virtual machines (SPICE clients), RDP clients, VNC clients, nested kwin/gnome-shell/whatever. See also: https://bugzilla.redhat.com/show_bug.cgi?id=1285770 . Games might also be affected, I don't know.
I'm posting this here to raise attention because
1. to fix this we need another protocol (extension)
2. this issue can't be fixed in a single application, it needs the whole stack (wayland library, compositors, affected applications) to change
Compared to the other bugs in "WaylandRelated" lists  , these bugs need architecture design, not just fixing bugs in one package.
I don't know whether this mailing list is the right place to discuss an issue like this. If you know better, please tell me.
Currently we have coredumpctl installed by default, but it doesn't work
because core dumps are consumed by ABRT instead. I'd like to get
coredumpctl working out-of-the-box. We can do this by changing our
systemd presets  to disable abrt-ccpp.service.
ABRT will continue to function thanks to the ABRT developers' work on
coredumpctl integration via abrt-vmcore.service. There is a somewhat-
reduced feature set, but I've been running this for several months
happily and I haven't noticed any issues; my crashes are being reported
to FAF, and I can manually report to Bugzilla. See  for details.
This might be slightly controversial for two reasons. First, the big
change here is that core dumps will appear in the magical and super-
awesome coredumpctl tool in addition to ABRT, but would no longer be
created in the cwd when ulimit is set appropriately (we would need to
feature that in the release notes). This aligns us with upstream
systemd behavior, but diverges from Debian and Ubuntu, which still
create the old-style coredumps in the cwd. Second, my understanding is
that the ABRT developers prefer to improve abrt-cli and tell people to
use that rather than coredumpctl. But as coredumpctl is very mature
(nicer) and cross-distro so it gets many more contributors, I
definitely prefer to instruct people to use coredumpctl instead for
We could totally do this change for F24 as well as it should be just a
one-line change, but in the spirit of beta freeze I figure it's
probably best left for F25.
Recently the Workstation working group agreed to match the GNOME apps
we install by default in Fedora with upstream GNOME's core apps. We
will by default sync our default apps to upstream's, and make
exceptions only for exceptional reasons if when proposed on this list.
This adds a bit more cement to our status as the best GNOME
distribution with minimal divergence from upstream.
I worked upstream with designers and release team to move many
desirable apps into core, to better match what we're currently shipping
in Fedora. I also created a gnome-incubator metamodule upstream, which
is where we'll put apps that we want to be core apps by design, but
which we recognize are not yet ready to be installed by default in
distributions. Currently that is home to bijiben, gnome-boxes, gnome-
calendar, gnome-dictionary, gnome-music, and gnome-photos.
Here is how we currently diverge from the new upstream recommendations:
Apps missing from Fedora: epiphany, gnome-logs
Extra apps in Fedora: bijiben, evolution, gnome-boxes, shotwell, rhythmbox
I see little value in discussing Epiphany again right now. Let's make an exception for that.
I have added Logs in Fedora just now, as I expect that will be uncontroversial and I don't see any value in diverging from upstream here.
Bijiben is just not very good yet, and it's not under active
development. There's no good reason for us to diverge from upstream
here, and I expect it will be uncontroversial, so I've dropped it.
I guess Evolution might be controversial; I use it religiously, and
many of you probably do too. But the user interface is complex and
confusing; users should not be exposed to this by default, barring
drastic UI changes that are outside the scope of the Evolution project.
Evolution is a great mail client for power users, but I'm confident
that the average user will be better off with webmail services; folks
who want a desktop client can simply install one, after all. We intend
to replace it with GNOME Mail eventually, but nobody has started
developing it yet. I propose we drop Evolution, but I have not done so
yet, pending further discussion.
Boxes has too many serious bugs right now, so it cannot go into gnome-
core yet, but we intend it to eventually. Since this is a significant
application, I have not yet removed it from our default install,
pending further discussion. My main concern is that it would look odd
for us to remove such a significant app from the default install, then
bring it back in a year or two. On the other hand, users won't notice a
thing unless they make a habit of reinstalling Fedora, and it's not
good to include apps we can't fully recommend. It's not clear to me
what choice is best here.
I propose we make temporary exceptions to keep Shotwell and Rhythmbox, until their intended replacements, Photos and Music, move into
upstream's GNOME core moduleset. That will very likely happen for
Photos for F25. I'm less certain about Music.
We've discussed replacing Shotwell many times in the past. I believe
consensus was that gThumb would be better, but we would instead wait
for GNOME Photos to mature a bit more so we don't change the default
photo app so many times. Photos is pretty clearly more polished than
Shotwell at this point, and I think we're ready to make the switch. Of
course it is still under active development, but let's not make perfect
the enemy of the good. If you care about this change, please either +1
Note that Image Viewer (eog) would still be the default app for opening
images, as it already is. That's not changing at this time.
P.S. As I've sort of become the "default app guy," I'll mention that I
have no further changes proposed for F25. The only other change for F25
(made a few months ago) was to drop Vinagre in favor of GNOME Boxes.
Tuned is an excellent tool to monitoring system and apply profile-based
configurations. This tool is very powerful and its use is relegated to
the servers systems, however, it can be put to good use for fedora
Workstation for users seeking advanced settings via a user interface
For example, a good use of tuned in the Workstation is the advanced
power management, even with the possibility of using custom profiles
through powerTOP (laptop users). My experience says that in most cases I
have used systems tuned for that purpose, there has been an increase in
battery life (is not the only use case).
Part of my proposal considers:
• Improve the design of tuned-gtk (like gnome-tweak-tool design)
• Add categories to target profiles to facilitate user choice (Eg.
Workstation, Cloud and Server)
• Add/modify new default tuned profiles for desktop/laptop use, similar
to those available in the "tuned-profiles-compat" package
• Add settings to create a powerTOP profile or add it from a report
All these options would expect for an average user or an advanced user
to enhance the options available for their device.
I intend to present it in the list and not directly to the authors, is
that this tool can greatly improve the user experience for the target
audience of fedora workstation. For example, the web is full of requests
"to improve power consumption in Linux" and there are several responses
that lead to programs that do not have a user interface and are
complicated to use. Tuned-gtk would be used for the range of user need
only choose a predefined profile to those who want to create their own
custom profiles graphically.
When I talk about advanced users, I'm making the parallel with the
application gnome-tweak-tool, but the big problem here is that the user
needs root permissions to use the application, although a change in this
area would help the whole family derived from fedora, as RHEL or CentOS.
Due to Flock next week in Kraków, I won't be able to chair the
scheduled Workstation WG meeting on 2016-Aug-03 at 10am US-EDT (1400
UTC). We have at least two options:
* Someone else chairs, volunteers welcome
* This might be better, if lots of other WG members are also at
Paul W. Frields http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
http://redhat.com/ - - - - http://pfrields.fedorapeople.org/
The open source story continues to grow: http://opensource.com
I have been working on a task related to this feature https://fedoraproject.org/wiki/Features/GConfRemoval, which is marked as 90% done.
I created a pulseaudio plugin called module-gsettings that mimics the behaviour of the module-gconf module. I already submitted the patch to the pulseaudio mailing list. The maintainer seems to be ok with adding it, with some comments that I tried to address. It adds a gschema for pulseaudio modules.
The main user of the old module-gconf is 'paprefs' which is unmaintained but is probably still used. I therefore created a patch for paprefs to move from gconf to gsettings. I did it so that a .convert file can be used to convert existing gconf configuration to gsettings. I could not find any way to submit my patch to paprefs though :-/
The pulseaudio patch is available here : https://github.com/lebauce/pulseaudio/tree/module-gsettings
The paprefs patch is available here : https://github.com/lebauce/paprefs/tree/master
Any comment, review, advice, gsettings knowledge, test would be more than welcome !
And if you know how to submit patch to paprefs, please let me know otherwise, I'll ping the last persons that commited to it.