I remember that the WMs situation was far worse regarding ICCCM. Bad behavior between the WM and the apps was usual. So I am really not choked that the development of a complex feature like CSD generate some bugs.

Of course only projects that add features are likely to add bugs. If you are perfectly happy with a specific conservative distribution, then just use it. But don't ask motivated developers to not innovate.

I still fail to understand why you want to use systemd-logind with an other init system. It's like trying to use one of the postfix binary without postfix infrastructure. I can't get it.

The client side decoration transition will certainly not limited to Gnome. KDE is also investigating something in that direction and will is likely to trig similar kind of bugs someday.

My personal opinion is that CSD has a future only is the WM has a very good GUI to manage the applications. This is the trend in most smartphone GUI and this is for reason: CSD is space efficient, but SSD is more reliable.

The twos bugs you linked was fixed.

On a embedded system I have build using Debian Jessie, systemd is very verbose, more than sysvinit in my opinion. I wonder if the fact that systemd don't say so much in your case is not the result of a Fedora choice to make systemd less verbose by default.

https://wiki.debian.org/system...

The curious thing is that the systemd binary only read unit files and execute command accordingly. How can it bug because of a USB device?

I agree with you. Some project take time to get accepted and every bug is potentially a drama. This is nothing new, some venerable UNIX projects have also pass trough similar problems. We tend to forget this because there are now regarded as good old stable, but this was not always the case when there was published.

You can perfectly run Gnome applications on a KDE desktop as you can perfectly run KDE applications on a Gnome desktop. If you can prove the contrary, please report the problem.

And why did you complain on systemd not displaying something if the problem is completely unrelated to systemd (it could be a kernel/udev issue)? Now I agree that something is really wrong if you can't at least get a text login console, but again if it's a kernel hand, systemd can't magically bring something on the screen.

I don't say that it impossible that systemd is in fine the cause of your problem and that fix is needed to solve your case. But without more information is't both impossible to clearly prove that's a systemd bug, and impossible to create a fix if it's the case.

I never say to report directly to systemd project. Please report first to the Fedora community.

I don't know Fedora, but on Debian you can install sysvinit or upstart as a systemd alternative to help testing if the bug also show without systemd.

Did you understand that you compare a file manager based on a web browser to a login deamon???

Anyway systemd-logind uses the systemd API and Konqeror uses the KDE API (many of them actually) and Qt API. What's the problem? The dependency on the respective API is no more or less on the two example. Also, the two applications rely on D-bus and so indirectly depend on the services there expect to use on D-bus.

https://packages.debian.org/je...

The Gnome 3 windowing issues have nothing related to systemd architecture (and I personally prefer XFCE or MATE).

From what I remember from the time when dbus was created, the main issue at that time was that there was too many different messaging systems and, in addition to some technical issues, none was gaining enough consensus to normalize the situation. Both Gnome and KDE have experienced how difficult it was to change the messaging system only to found that there are each duplicating many works in a incompatible way.

The requirement to define a strict binary protocol was found to be the best way to normalize the situation. And this was a big success: since dbus is used by Gnome and KDE, all the low level services can publish an API that can be used by both projects. The initscripts didn't follow this trend and are now regarded as obsolete because there are unmanageable from dbus. I hope this show how normal is the central position of dbus into systemd.

The fact is that you want a machine that boot and automatically start some services. Yes you can manually call initscripts, but you need an other part to make you scripts called at boot time. This other part is either sysvinit+sysv-rc, or sysvinit+file-rc, or upstart, or systemd, or some others unusual ways. So I still keep my position that's the initscript that depend on sysv-rc or something other to be called at boot time. it's not an hazard if the Debian dependency graph has been corrected that way when upstart was packaged.

Calling a script at boot time with systemd is no a hack at all, and from the script point of view this will change nothing.

I don't understand how the konqueror example is related to systemd.

Ok, I understand you position.

Actually dbus is essentially a normalized binary protocol (http://dbus.freedesktop.org/doc/dbus-specification.html) and this was on purpose to allow differents implementations to cooperate. The communication part (the transport in dbus jargon) is not so strongly defined (there is multiple transports). How to send and receive a message could see some evolutions in the future (for example using kernel dedicated message passing instead of general sockets).

The idea is to not bound developers to a specific code to exchange dbus messages. From the view of the others applications, a specific application only have to comply to the binary protocol and to specify an API.

Nice to see that you make real analysis of systemd. I have read your progress.

I found curious your position about dbus. Since it's a central piece since so many years, it's logic that more and more services depend on it to publish an API.

Have you considered that your various reasons to disagree on systemd regarding UNIX philosophy could be useful to others peoples?