GNOME Shell 3.10 Is Ready To Shine On Wayland

[intellivision]

well i guess in some months you won't have that issue since i expect most distros and major Desktop Enviroments to use systemd anyway and the remant bash warriors will find some way to fake the dbus api using bash/perl. Gnome/KDE/Etc should set as dependency the tools they consider best for their project and distros should be the ones that make sure to fulfill those requirements, if your favorite distro is not able to do so, well is a free world go and choose another DE that works nobody is gonna miss you

Well Gnome is the only DE that has gone against the grain and used systemd exclusively. I doubt KDE or any of the other main desktop environments would be stupid enough to alienate their users without any upgrade path.

[jrch2k8]

You mean except the largest distros (Ubuntu, ChromeOS, Android, for Debian we still don't know) and Gnome. For KDE I don't see a move to systemd, neither for XFCE or any other DE.

You will, once you realized that eliminating choice and depending on only one software is a bad thing for Linux. But of course you will deny that now. As I already told that Honton troll: If you are against choice and find comfort with that maybe try Apple, seems to fit.

well KF5 is on heavy development but already the idea of burning with fire KDM codebase entirely is there and the logind integration is there too, check the mailing list is actually free

Systemd is not a problem of use this by force because evil lennart and redhat wanna control the world, the problem lies that systemd is already too far away from any other init system in features that for developers is like go back to use wooden square wheels in their cars after have tried solid bullet proof round wheels with radials.

the problem is not systemd, is all other systems that refuse to leave the 90's SCO era

systemd for users completely destroy any other choice in features and for developers[especially ISV and DE teams] it make things 100000000 times easier, faster, more reliable and standarized across distro(ofc you have to bother in do the mortal sin of going to http://www.freedesktop.org/wiki/Software/systemd/ and do the heretic act of read a bit )

[jrch2k8]

Well Gnome is the only DE that has gone against the grain and used systemd exclusively. I doubt KDE or any of the other main desktop environments would be stupid enough to alienate their users without any upgrade path.

well KDE 4.11 already got as heavy dep polkit[same as gnome 3.6 i think] and it already uses heavily logind through polkit too, the proper integration would be expected in 5.0 or 5.1 release since KDM is getting totally removed and in next kernels/polkit VT_CONFIG/consolekit support in polkit are finally taking their retirement package.

E18 should be wayland/systemd combo, xfce well after gtk3 migration dunno which session manager they will use, cinammon/mate i think they relay on gdm, etc

[Vim_User]

well KF5 is on heavy development but already the idea of burning with fire KDM codebase entirely is there and the logind integration is there too, check the mailing list is actually free

Systemd is not a problem of use this by force because evil lennart and redhat wanna control the world, the problem lies that systemd is already too far away from any other init system in features that for developers is like go back to use wooden square wheels in their cars after have tried solid bullet proof round wheels with radials.

the problem is not systemd, is all other systems that refuse to leave the 90's SCO era

systemd for users completely destroy any other choice in features and for developers[especially ISV and DE teams] it make things 100000000 times easier, faster, more reliable and standarized across distro(ofc you have to bother in do the mortal sin of going to http://www.freedesktop.org/wiki/Software/systemd/ and do the heretic act of read a bit )

You can be assured that I have "read a bit" and actually tested that abomination in Arch. Was anything but a good experience.
Anyways, the reasons why systemd is forced into the market (IMHO it is rather the "Hey, you can use anything modular, oh wait, you can't we just added a few dependencies in the last version" approach than superior features) are totally irrelevant, it doesn't change a bit that it is a very bad idea to put all your eggs into one basket. You will neither be the first nor the last person to find such elemental knowledge that applies to anything in life, but sooner or later you will find it.
When the point comes that every distro has to use systemd Linux is officially dead for me and I have to look for sane alternatives.

[intellivision]

well KDE 4.11 already got as heavy dep polkit[same as gnome 3.6 i think] and it already uses heavily logind through polkit too, the proper integration would be expected in 5.0 or 5.1 release since KDM is getting totally removed and in next kernels/polkit VT_CONFIG/consolekit support in polkit are finally taking their retirement package.

E18 should be wayland/systemd combo, xfce well after gtk3 migration dunno which session manager they will use, cinammon/mate i think they relay on gdm, etc

So what you're trying to say is that Gnome is the only desktop environment to make the inept decision to support only logind without any alternative.
Good to see that you've reiterated my point there.

[mrugiero]

You can be assured that I have "read a bit" and actually tested that abomination in Arch. Was anything but a good experience.
Anyways, the reasons why systemd is forced into the market (IMHO it is rather the "Hey, you can use anything modular, oh wait, you can't we just added a few dependencies in the last version" approach than superior features) are totally irrelevant, it doesn't change a bit that it is a very bad idea to put all your eggs into one basket. You will neither be the first nor the last person to find such elemental knowledge that applies to anything in life, but sooner or later you will find it.
When the point comes that every distro has to use systemd Linux is officially dead for me and I have to look for sane alternatives.

Then don't. If you think putting all your eggs on a basket is a bad choice, implement features somewhere else, instead of trying to force developers to tie things with wires everywhere.

[intellivision]

Then don't. If you think putting all your eggs on a basket is a bad choice, implement features somewhere else, instead of trying to force developers to tie things with wires everywhere.

They weren't complaining about it before systemd came along and tried to assimilate the Linux space.
And it looks as though the largest players such as Google and Canonical will not be moving their platforms' bases to systemd either, so now the workload of porting Gnome to those platforms has increased exponentially.

[mrugiero]

They weren't complaining about it before systemd came along and tried to assimilate the Linux space.
And it looks as though the largest players such as Google and Canonical will not be moving their platforms' bases to systemd either, so now the workload of porting Gnome to those platforms has increased exponentially.

Where "increased exponentially" equals writing a backend to upstart or whatever they want to use for the needed features, if it provides the features expected. Even more, since there was a ConsoleKit backend for the same features (which bitrot, because nobody was willing to maintain it when they called for it), I'm pretty sure the work on the abstraction front is already done. Only thing lacking is the actual backend.
Another thing is Canonical uses logind already (they froze to an older version, yes), and I bet Google uses the same.
I'd expect them to switch as soon as Debian switches.

Also, before GUIs existed, there was no one complaining they didn't have one. But try to give a random user a CLI, and look at the reaction.

[intellivision]

Where "increased exponentially" equals writing a backend to upstart or whatever they want to use for the needed features, if it provides the features expected. Even more, since there was a ConsoleKit backend for the same features (which bitrot, because nobody was willing to maintain it when they called for it), I'm pretty sure the work on the abstraction front is already done. Only thing lacking is the actual backend.
Another thing is Canonical uses logind already (they froze to an older version, yes), and I bet Google uses the same.
I'd expect them to switch as soon as Debian switches.

So if the work isn't that hard to undertake, why hasn't the Gnome team done their job and supported those setups which thousands of people use every day?
Also, Ubuntu may support a legacy version of logind but that version won't be compatible with 3.10, so Gnome users will be left behind by their own upstream on Ubuntu as well.
Furthermore, I'd like to see this proof that Google does the same with their Linux OS's Chrome OS and Android, unless you state that what you've said is pure unsubstantiated conjecture.

[mrugiero]

So if the work isn't that hard to undertake, why hasn't the Gnome team done their job and supported those setups which thousands of people use every day?
Also, Ubuntu may support a legacy version of logind but that version won't be compatible with 3.10, so Gnome users will be left behind by their own upstream on Ubuntu as well.
Furthermore, I'd like to see this proof that Google does the same with their Linux OS's Chrome OS and Android, unless you state that what you've said is pure unsubstantiated conjecture.

I have no proof. I just don't see them using something different than Ubuntu for Chrome OS, since AFAIK is a modded Ubuntu with a custom DE. Getting picky means extra maintenance for something it's not worth. I don't think there is a sizable amount of Android users interested in full fledged DEs, so I don't care about Android. And again, I haven't said they DOES, I said I expect them to. Those are two very different things.

On another front, developers in the free software world are either paid by someone interested in something (Red Hat will not pay support for something their users don't have, and guess what Red Hat distributions don't use) or work for free, because of their own interest. On neither case should you expect to support anything they or the ones paying them don't use. They accept patches, they do code revision, and that's as far as their legitimate responsibility with anyone else goes. You want a feature? Great, write a patch. It will go through revision, and if it doesn't screw anything it will probably get accepted. That's how it works. You are not a coder? Well, then, you can start a kickstarter (or directly pay up), and ask someone to code for you in exchange of money. Else, quit giving orders, you are nobody's boss.