Future Of Fedora Spins Is Questioned With Fedora.Next

[Sonadow]

I have been a Red Hat /Fedora user since RH 4.2. I like Fedora and its pace, and on the whole been satisfied with its direction, except of course that I use KDE instead of the Gnome.

But there is something incredible creepy about the entire "Fedora.next" process. It is all top down steering, with vague and unclear goals.

I may be wrong about this, but there seems to be a underlining sub text in what is going on right now, that Gnome will be the only official, tier one supported desktop, with other DE's hidden way and dropped to "barely tolerated" support.

So if the Fedora.next project effectively is going to nerf KDE, I will be leaving.

Does it _even_ matter?

Fedora has shown no signs of intending to drop KDE from its repositories, so there will always be an up-to-date version of it available for install.

[Sonadow]

That 5 minute edit limit...

Does it _even_ matter?

Fedora has shown no signs of intending to drop KDE from its repositories, so there will always be an up-to-date version of it available for install.

What i wanted to say was, even if KDE does get relegated to a tier 2 or tier 3 or tier X desktop, as long as there exists a working and up-to-date version of it in the repositories for installation, it's already good enough.

[interested]

That 5 minute edit limit...



What i wanted to say was, even if KDE does get relegated to a tier 2 or tier 3 or tier X desktop, as long as there exists a working and up-to-date version of it in the repositories for installation, it's already good enough.

Well, that is the situation right now; KDE is a second tier DE, but thanks to the excellent Fedora-KDE release team, it is a good DE experience to use, and the KDE spin makes it easy to choose and install too.

The problem may be, that "Fedora.next - Workstation" will be defined as a Gnome only, with the ultimate goal of becoming _the_ "GNOME OS" implementation.

KDE, LXDE etc. could then be relegated to "slum repos" that are "community driven" in the sense that basically all official support will be removed, perhaps bug-trackers too, and they will no longer be part of any Fedora release schedule.

KDE, LXDE, Xfce would basically be banished into Fedora Remixes. So no spins, no official support or QA for KDE and other DE's.

I am not saying it will end up like this, but there sure are some signals in that direction on the Fedora project mailing list.

[Awesomeness]

So if the Fedora.next project effectively is going to nerf KDE, I will be leaving.

Won't happen. If that was the case, Red Hat would not pay so many KDE devs. They'd be all fired already.

[interested]

Won't happen. If that was the case, Red Hat would not pay so many KDE devs. They'd be all fired already.

Well, there is no Fedora.next yet. AFAIK there are 4-5 KDE developers working for Red Hat, not that many. Don't even know if they are full time employees or not.

But as the Fedora.next "Workstation" proposal look, then it will a set of mandatory packages. That looks like "Gnome only" to me as things are now, and with the disbanding of the spins, there will not be a second "Fedora KDE Workstation" platform. One solution for Red Hat could be that they banish all other DE's into "Fedora Remixes", basically removing them from Fedora.

If all the GNOME OS hyperbole about Gnome getting 20% of all desktops in 2020 is believed by Red Hat, then it will become a natural choice to remove all Gnome competitors from their distros.

The mailing list and the fedora.next proposals are full of talk about unification, standardisation, uniform experience and similar stuff, that looks like they are preparing for only one official DE, Gnome. Perhaps they think that Ubuntu's success is down to only having Unity.

[zxy_thf]

I have been a Red Hat /Fedora user since RH 4.2. I like Fedora and its pace, and on the whole been satisfied with its direction, except of course that I use KDE instead of the Gnome.

But there is something incredible creepy about the entire "Fedora.next" process. It is all top down steering, with vague and unclear goals.

I may be wrong about this, but there seems to be a underlining sub text in what is going on right now, that Gnome will be the only official, tier one supported desktop, with other DE's hidden way and dropped to "barely tolerated" support.

So if the Fedora.next project effectively is going to nerf KDE, I will be leaving.

imo Fedora.next makes SIGs working independently (or "modularized").
I think I can expect receiving a major DE update within several weeks after it being released in Fedora.next, rather than several months.

[RahulSundaram]

Well, there is no Fedora.next yet. AFAIK there are 4-5 KDE developers working for Red Hat, not that many. Don't even know if they are full time employees or not.

But as the Fedora.next "Workstation" proposal look, then it will a set of mandatory packages

There are more than 4 to 5 people working on KDE *in* Fedora and it is a mix of full time Red Hat contributors (status reports at http://ltinkl.blogspot.com/) and several volunteer Fedora contributors who have been associated with the project for a very long time. Workstation is just one of the products in the proposal and it doesn't make sense to talk much mandatory packages when you could always do a network installation and select any group of packages they need or just ignore the products and use kickstart to cherry pick and control the set of packages to an artibrary set of packages.

There are no plans to drop any desktop environments off the repository. That is just wild speculation on your part. The current discussions are solely about the best way to deliver them.

[interested]

There are more than 4 to 5 people working on KDE *in* Fedora and it is a mix of full time Red Hat contributors (status reports at http://ltinkl.blogspot.com/) and several volunteer Fedora contributors who have been associated with the project for a very long time. Workstation is just one of the products in the proposal and it doesn't make sense to talk much mandatory packages when you could always do a network installation and select any group of packages they need or just ignore the products and use kickstart to cherry pick and control the set of packages to an artibrary set of packages.

There are no plans to drop any desktop environments off the repository. That is just wild speculation on your part. The current discussions are solely about the best way to deliver them.

Workstation may only be one "product" of several, but if it is the only desktop image available on "get fedora", and it is strictly Gnome only, then KDE and all other DE's will receive a solid kick down the ladder. It will be a clear signal that Gnome will be Fedora.next's primary DE, and all others are merely tolerated, and perhaps only for a while.

I know that you aren't speaking officially for RH, and that Fedora.next is still very vague and unclear, but as I understand the your answer, you too seem to acknowledge that the Fedora.next Workstation "product" will be strictly Gnome only. You just don't see it as a problem, since one can use the net installer. Well, allow me as a long time Red Hat and Fedora user to strongly disagree with that. And seriously, alienating the KDE, LXDE, and Xfce users isn't the way forward for Fedora to gain new users and developers.

[eidolon]

While downloads through fedoraproject.org don't tell the whole story for reasons that are commonly mentioned when trying to approximate actual usage, does the Fedora Project track downloads in such a way it can be determined which install media are most downloaded (e.g. DVD, netinstall, KDE live media, GNOME live media, Xfce live media, etc)? I'm going to hazard a guess that GNOME live media is the most downloaded, at least in part due to its being the Fedora default and relatedly the most visible download. I know that spins.fedoraproject.org ranks spins by popularity, though there are no numbers provided to put that in context, nor does it shed any light on downloads for DVD, netinstall, or GNOME live media. That (i.e. a piece of the puzzle of how Fedora gets into the hands of users) may be completely extraneous to the Fedora Project's current internal discussions, but I would like to know if download figures are collected somewhere, and are readily accessible (internally or externally to the Fedora Project).

[Awesomeness]

tl;dr: Korora can only win.

AFAIK there are 4-5 KDE developers working for Red Hat, not that many. Don't even know if they are full time employees or not.

AFAIK only Blue Systems has a similar number these days. ?In the good ol' days? SUSE had ~20 KDE developers but these either no longer work for SUSE or have been reassigned to work on other projects. If they are active in KDE at all, they do it in their spare time.

The situation right now is kinda paradox: Fedora ships Gnome by default and Plasma Desktop is only second tier but Red Hat directly employs KDE developers, whereas openSUSE ships Plasma Desktop by default but SUSE no longer employs developers to work on KDE software. Heck, apparently SUSE can't even fix bugs in the FOSS GPU drivers openSUSE ships. Months ago I filed a report against a bug in Nouveau in openSUSE 13.1 that left my NVidia-powered notebook unbootable. Nobody ever looked at that bug report. On my new laptop (Intel integrated GPU and AMD dedicated GPU) all distributions I tried (openSUSE, Ubuntu, Debian,?) booted right into a black screen ? all except Fedora. I find YaST and the ?fork me? mentality on build.opensuse.org to be openSUSE killer features but I prefer a well supported second tier over a badly supported first tier. If only Fedora works on my hardware without hacking, I use Fedora and Fedora.Next won't change that.

But as the Fedora.next "Workstation" proposal look, then it will a set of mandatory packages. That looks like "Gnome only" to me as things are now

Right now the ?Desktop? is Gnome-only (there is no ?Gnome? variant of Fedora, Fedora Desktop simply ships Gnome). However, the KDE variant AFAIK is the only Spin with the power to block Fedora release, i.e. if there is a showstopping bug in a crucial KDE / Plasma Desktop component (KWin for example), a Fedora release won't ship until that bug is resolved.
No other desktop has that power. Not Xfce, not LXDE, not Mate, not Cinnamon. And that's why Fedora ships Plasma Workspaces as default on ARM.

If anything, the current Fedora KDE spin will likely be promoted from a mere Spin to a Fedora Product. As for the other desktops: Their fate is not so sure but I doubt they'd be eliminated from the repos.
Personally, I think as long as they are kept in the repos, delegating responsibility for default setups to remixes is not that bad. I do not use default Fedora. I use the Korora remix. Korora is great and if Korora gets more volunteers by becoming the number one destination for people who want to have Plasma Desktop, Xfce, LXDE or whatever by default, it'll be great.

If all the GNOME OS hyperbole about Gnome getting 20% of all desktops in 2020 is believed by Red Hat, then it will become a natural choice to remove all Gnome competitors from their distros.

In RHEL Gnome and Plasma Desktop are both fully supported by Red Hat. AFAIK other desktops aren't even in the official RHEL repos but KDE/Plasma will be fine.