addresses a distinct, relevant, and broad use-case or user-base that a Fedora Edition is not currently serving;
is a long term investment for the Fedora Project; and
is consistent with all of Fedora’s Four Foundations.
We propose an additional line:
The Council may make exceptions to the “distinct” rule when we determine that doing so best fits the Project’s Mission and Vision.
This change reflects actual practice. For example, there is clear overlap between the use-cases and potential user-bases for the current Server, Cloud, and CoreOS Editions, but each takes a different approach. We are currently considering an exception for a KDE Desktop Edition, which would overlap with Fedora Workstation.
This topic is open for community discussion, following our Policy Change Policy. After two weeks, the Council will vote in a new ticket, and if approved, the policy will be updated.
Approval of this change would not automatically mean that approval of a KDE Desktop Edition, but would allow that possibility.
I’m in favor of the substance of this proposal, but I don’t like the implementation. I’d propose, instead, to change the first point to
addresses a sufficiently distinct, relevant, and broad use-case or user-base that a Fedora Edition is not currently serving;
This gives the Council some latitude without creating a “but we can choose to ignore this when we need to” carve out. It gives the Council space to determine what is “sufficient” for a particular proposal. It’s a big vague, but no more vague than the existing text is in practice.
This also means the Council could choose to promote a deliverable to Edition when the use case or user base is narrow-but-important without having to go back and add yet another exemption. For example (and I am not proposing this, to be clear), if there’s a Fedora LLM Edition that has few users but is considered fundamental to the future of the proejct.
If we were going to wordsmith a rules change, I’d just drop the “that a Fedora Edition is not currently serving” part. That acknowledges the reality that Editions are not actually based on this, but instead based on community desire and engagement.
I think that opens the gates a little too wide for what I want to do right now. We might want to consider that in the future (it’s true that community desire and engagement ultimately drive everything), but what I want to do right now is:
Bring our policies minimally in line with what exists today, and
Give ourselves the option of considering another desktop Edition.
Because we do treat Editions differently, there is a cost across the project to each edition. Quality, rel-eng, packagers, marketing, design, website, docs, Ask Fedora, and other teams are all asked to take on more. When a new Edition overlaps with an existing one (or changes to an Edition or in the world create an overlap between two existing Editions), that has a cost too. We want a family of Editions that support each other, not accidental zero-sum games.
I do like Ben’s elegant one-word change better than my exception text, though.
In practice, it doesn’t because every Edition has to be approved by the Council anyway. There’s no automatic mechanism to create an Edition, it has to go through a review anyway.
It could be that English isn’t my mother-tongue, or it could be my personal experience, but am I the only one that reads this as already (at least somewhat) compatible with a second Edition, with no need of modification?
I would have personally migrated to another distro after my first month of Fedora usage, because realistically, Gnome did not, indeed, serve me. KDE kept me around.
If there’s more like me, then there’s a user-base, which is what the definition specifies, no?
To me, the distinction appears to then be:
Which is something the council can directly decide if is true or not for any “Remix” that becomes ambitious of Edition Status.
Yeah, but it’s good to have a written-down set of basic criteria.
I see what you’re getting at, but I don’t think we’d generally accept “people who prefer a particular desktop environment” as a target user base. Our desktop target is really something like “general computer users, with a special but not exclusionary focus on Linux enthusiasts and future possible Linux enthusiasts.” (That’s not official or anything, it’s just kind of where things seem to be practically after 10 years.)
My opinion is that there’s room in that target audience to provide two desktop environments, but not that they are really different cases. Does that make sense?
Totally support! I would love to see Plasma promoted to a full edition instead of a spin. It’s been a release blocker (The only other one besides GNOME, iirc!) for ages, and it richly deserves to be featured more prominently as an honest-to-goodness edition addition. Policy change looks good too!
In short, I am +1 for the proposed change, and add that I like the suggestion of @bcotton.
I am neither +1 nor -1 about the idea of @ngompa because, as he said, it is anyway the Council that makes the decision and derives this decision from the rules, whereas “that a Fedora Edition is not currently serving” can be construed anyway in many directions. I don’t think this part can hinder a new edition if people want it and the Council wants it as well, whereas I thus also agree that the part does not add much value because it is too generic and abstract to cause a real limitation. It is other rules that limit new editions or not. So I’m fine with and without it.
Anyway, I would be happy if that change later facilitates a KDE edition. Thanks to the Council and the KDE SIG for their efforts!
I wonder if in addition to loosening the policy for Edition promotion as per current proposal, there should also be an uncorrelated tightening as well, i.e. requesting for necessary resources to be identified in order to keep the new Edition at the level of expected standards.
There is no win in promoting a spin to Edition status, if that in turn reduces the quality of existing Editions as a result of limited resources being shared among too many projects.
every sitiation has there u and down sides, but in this case i see more ups. Bringing KDE as edition will benefit users and Fedora lot more than just current editions.
As i see this is users an choose oficial edition as they want, Bring more users, Contributers in long run. i Think this is long run benefitical for Fedora as there is 2 difrent DE editions testing can be more resource heavy at first, but lets say example F41 had on gnome where apps didint start, but other hand KDE didint had issues so this will help debugging issues too to narrow down issues is those Gnome or KDE or Fedora side issues
Note that this policy change is not actually intended to block promoting the KDE Desktop Edition. That is not how we discussed it in the Council meeting either. So this is not blocking promotion, but rather changes the way the promotion occurs.
I am in favor of the proposed policy change. I agree that the current policy does not reflect current practice, and have always considered it an unfair impediment to alternative technologies (and in particular alternative desktop environments) to begin with.