#1243 Consider release blocking status of KDE spin(?) for Fedora 21 in .next decision-making
Closed None Opened 5 years ago by adamwill.

Wasn't sure whether this fit neatly into any of the other .next tickets, so thought it best to file a new one.

We (QA) agreed at our weekly meeting that it'd be good for FESCo to explicitly consider the release-blocking status of KDE per se and as a deliverable for Fedora 21, as part of the .next planning. At present this status isn't entirely clear. See http://meetbot.fedoraproject.org/fedora-meeting/2014-03-03/fedora-qa.2014-03-03-16.02.log.html . Thanks!


The summary of the Workstation product is somewhat incorrect, but that's mostly because we haven't been overly clear. There have been members of the WG that have suggested we make KDE a release blocking desktop for the Workstation product. How exactly that is done is unknown at this time, but the QA meeting minutes are correct in that there won't be a specific/separate KDE image.

Off the top of my head, the KDE testing would be something along the lines of "install the Workstation live image, install KDE through software-installer (if necessary), log into KDE from GDM after install, test". I'll try and follow up on KDE this week. If that (or something similar to it) is the approach, would QA consider that to be sufficient for release blocking status? Or would you still test a KDE spin from the KDE sig?

https://lists.fedoraproject.org/pipermail/desktop/2014-March/009458.html for the thread I said I would follow up on (clearly no replies yet as I just sent it).

I've always said that my take is that that kind of question is really a policy question; it's not for QA to decide. It's more - do we think that's an appropriate deployment method for KDE, for the Fedora project? Are the KDE folks happy with it? Are the desktop/Workstation folks happy with it? Is FESCo happy with it? If the project as a whole (particularly as channelled through those groups) thinks 'deploy Workstation, then deploy KDE on top of it' should be our primary KDE deployment method for F21, and that KDE is still significant enough to block releases...that's really the question to answer.

Replying to [comment:2 jwboyer]:

https://lists.fedoraproject.org/pipermail/desktop/2014-March/009458.html for the thread I said I would follow up on (clearly no replies yet as I just sent it).

Replied to the thread, also added KDE list on CC. It's not about blocking status, more about how we want to deliver it.

I believe the following position to be shared by KDE SIG as a whole:

The Plasma Desktop (i.e. the KDE live image) really needs to stay
release-blocking, not just the runtime stuff on the GNOME image (assuming it
gets added at all)! There's no way the KDE SIG is going to compromise on
that.

We need no less than what we have now, a release-blocking Spin.

If you need official confirmation of that, I can put it up for a vote in the KDE SIG meeting next Tuesday.

And I personally don't see why it shouldn't be a Product in its own right, but this is still being discussed in the SIG.

See also: https://lists.fedoraproject.org/pipermail/kde/2014-March/013184.html

I wholeheartedly endorse kkofler's position here:

1) Yes, we want a KDE spin with every Fedora release
2) Yes, we want blocker status

From the 2014-03-05 FESCo meeting:

AGREED: defer this one week waiting for more feedback from both KDE SIG and Workstation WG (+:7,-:0,0:0) (notting, 19:25:45)

Right on the "About Fedora" page, it says "We don't mind shaking up the status quo, when it means we can more effectively move free software forward."

While in this instance, keeping KDE as a release-blocking spin of Fedora ''is'' the status quo, by removing this status we would no longer be moving "free software forward".

I arrived to being a KDE user the long way - I started with Windows 8, hated it, and moved back to Fedora. I was "introduced" to gnome 3 at that point, and unless you've just been using Windows 8 you probably wouldn't notice the stunning similarities. It was enough to make me dislike gnome3, also. I tried Cinnamon and Mate, none of which did what I needed them to do. I then decided to give KDE another shot, after a disastrous attempt years ago with KDE 2. To my sheer and utter surprise, it worked - and worked well. KDE deserves to be just as big a part of the official Fedora.next release as gnome does.

gnome isn't the answer - I'm sorry, but the direction they are going is not the direction Fedora.next needs to go. Fedora.next needs to offer - above all - stability, and secondly, choice. I may be stepping outside the bounds here a bit, but I would be tempted to propose that there be a Fedora.next product, and a Fedora.kde product.

Sooooo, we decided to delay this in hopes of getting more feedback. I haven't seen much further discussion. Did that happen in areas I missed? Do we need to make a more active action item for someone from FESCo?

I spoke to some of the KDE guys about this yesterday. They are currently evaluating several approaches (that I am aware of) before they push hard on this:
1. KDE as the basis for the Fedora Scientific Product (obviously begging the question of how they get that approved).
1. KDE as a traditional live spin.
1. KDE as the supported graphical environment for the Server product. (This was only suggested to them yesterday, so I may be premature in mentioning it here.)

Are those ideas for F21 or for F22?

I notice that none of the ideas are simply "Provide an OS with the K Plasma Desktop as the graphical environment". Is that out of desire or because we (FESCo) haven't decided on whether we'd accept new Products with an overlapping audience?

Replying to [comment:12 toshio]:

Are those ideas for F21 or for F22?

I notice that none of the ideas are simply "Provide an OS with the K Plasma Desktop as the graphical environment". Is that out of desire or because we (FESCo) haven't decided on whether we'd accept new Products with an overlapping audience?

The latter, I think. While we haven't expressly stated that we wouldn't support those cases, we've definitely implied strongly that a new Product would have a better chance at acceptance if it targeted an audience that the existing Product definitions don't address.

Also, having lurked at the Fedora-KDE SIG meeting today, it is pretty clear now that they plan to come to the Board and FESCo about a KDE-based "Educational and Scientific" product they're calling Fedora Plasma.

For the record (since I'll be missing the FESCo meeting today), I think KDE is a significant driver of adoption in the distribution and I would vote -1 to any attempt to remove it as a blocking desktop.

That being said, I think it's up to QA to decide the configuration(s) for which they will test KDE.

As Stephen said - we are discussing our own product right now and we hope, we will be able to present it to FESCo and Board soon. As from previous discussion, it was recommended to avoid overlapping target audience. This product is going to aim on educational and scientific target audience (as Stephen pointed out, Scientific spin is pretty popular, KDE is used by this community etc.). FESCo agreed on that new products process will be set once the first product comes. I really think Fedora Plasma could be a great example - the team proved to be able to create supported, release blocking spin for a long time and have resources and continuity. Ideally, we still would like to make it for F21.

Regarding QA (and other teams) resources - we are willing to offer (or better we try to provide) our resources to help coverage of Fedora Plasma product.

AGREED: defer 1-2 weeks for KDE Product Proposal. If we don't approve that in two weeks we can revisit whether to make the KDE spin release-blocking. (+7,0,0)

Deferring another week.

Agreed on today's FESCo meeting: Recommend to the board that they consider [Fedora Plasma] for 22 ... For F21 we continue to consider the KDE Spin (or a renamed version) as release blocking. KDE Spin will continue to have a place on the download page for the Fedora Products. (+5-2)

Login to comment on this ticket.

Metadata