@ngompa filed this ticket, and my bot did not bridge it, so I’m doing this manually. And then will go kick the bot.
We talked about this briefly at today’s Council meeting. See the minutes.
In very short summary, we agreed that
there should be some kind of general policy — a rubric — about these kind of things
we shouldn’t use Fedora community money as a proxy patch for “Red Hat ought to care about this project but for some reason does not seem to”
KDE is (historically and currently!) a popular and important part of Fedora
From that, this list of next steps (of which this is item 1):
Create a linked discourse topic, summarize this a bit
Fix my bot?
Point the KDE SIG, Workstation WG, and Mobility SIG at the topic
Find someone to write that general rubric
But even while that’s a draft, kind of get a sense on where we think this should fit
Make a decision on the rubric
Use that to make a decision on KDE patronage in specific
With “Marie and Matthew talk to RH OSPO about funding and sponsorships for this and other projects and how it all fits together from that perspective.”.
On one hand, I like the idea. I have myself started to work with KDE Eco a bit, which we will also represent on our Fedora booth at the FrOSCon. I like the idea of bringing both communities closer together.
However, there is already cooperation, and both know from each other, and users engaging in one community tend to also know the other. If a user has already made the decision to use KDE, I would expect most users do not make their decision of the Linux distribution based upon the sorting on the KDE page. From that point of view (and because we do not intend direct revenue), I do not see the advantage of:
The remaining points that seem very interesting are
So, I think we might first talk about if and how we will use the seat and the events? And which opportunities it might create? I mean, being on the advisory board is not an end in itself, is it?
Concerning networking, do we need KDE to network with these companies?
Does the advisory board enable us to pass/get information/incentives to/from the KDE board which we cannot pass/get through the KDE SIG? (here are some information about the advisory board)
Of course it might be added that the 5000 € would not be “gone” but used for means from which Fedora profits as well: the money will be used to, indirectly, improve Fedora KDE & Kinoite.
My personal opinion is that we should talk about these questions at first.
Personally, I think the incentives/information that are intended and interesting for the board (which are not just technical incentives) are those of Red Hat and not Fedora. And the latter should not replace the first.
Edit: Just to put an alternative into discussion: how about using these resources to accelerate the KDE SIG towards a full sub-project? Maybe not just for fostering development but also to facilitate common efforts beyond (like KDE Eco; just as an example) with dedicated resources; I am not sure if the advisory board makes a big difference in the immediate cooperation between the communities, but a sub-project might consolidate resources that are more worth than pure money. @ngompa Would that make sense as well from your point of view? I am not deeply into the activities of the KDE SIG.
We may, yes. It also helps where we can coordinate our voices where it makes sense to drive KDE to drive and evolve the project in a particular direction. For example, we’d like to get Fedora KDE preloaded on machines. To do that, we need to drive interest in supporting KDE on preloads properly. The opportunity to interface with these companies directly make it easier to accomplish that goal.
KDE, e.V. is the umbrella organization for KDE Plasma upstream. The KDE SIG is part of the Fedora Project. The KDE Advisory Board is part of KDE. Us having a seat means the Fedora KDE SIG has a voice at the highest levels of KDE.
It’s not indirect in the slightest. The Fedora KDE SIG packages and delivers KDE software. This is done through two variants: the Fedora KDE spin and Fedora Kinoite. Investment to support the development of KDE software directly benefits us.
Nowhere am I talking about Red Hat. Red Hat is firmly invested in GNOME. They sit on the Advisory Board for the GNOME Foundation and have for decades. They helped create GNOME in the first place.
While it’s true that Red Hat is the legal entity giving the money, they are doing it on behalf of Fedora, and Fedora is the organization being represented here, and a member of the Fedora KDE SIG would take the seat.
For various reasons, this will never be possible. The KDE SIG cannot evolve past its current governance model because becoming a full subproject (a working group) would directly conflict with the Workstation WG.
From the marketing POV, I think this could make us gain more users, because it would be on the news (I will make sure to promote it everywhere). And at the end more users will attract more contributors (in a proportion that I don’t manage yet).
I know is not a little money, but if it get approved, we need to make several changes, starting on becoming KDE WG, and here is where I want to ask why you said:
And after this, the main website getfedora.org should pair Workstation and KDE, because it will become a main edition, not a Spin anymore. Or at least that’s what I’m thinking.
There is a big likeliness of us garnering more users, to not only KDE Plasma spin but to other editions as well when this happens - when we promote the same on our reworked websites.
I meant the same with indirect: we put the resources into KDE e.V., and KDE e.V. puts it into improvements of Fedora KDE/Kinoite. So absolutely agreed.
I think we should point out that Red Hat and Fedora are two different things, and the first is more than the downstream of the latter: the incentives Fedora can provide are not the same Red Hat can provide.
If I review the patron list and the advisory board tasks, it looks to me that this board is primarily to gather the non-technical perspectives of downstream stakeholders (vendor, user and customer expectations, non-technical developments, synergies and such), so the non-technical stakeholder perspectives that potentially remain hidden to developers (and vice versa). But the information I have is admittedly very abstract and I have not yet worked with the KDE organization. Are you sure that this is a board with a focus in which the Fedora KDE SIG fits?
There are some articles about the board from the time it was introduced, which indicate such a non-technical scope. Unfortunately, I find only German versions, such as 1 and 2. That’s why I’m asking. However, the question of who takes the seat in the advisory board might be separated from the question if patronage makes sense.
It sounds like Fedora would be buying our place in KDE community.
If we are talking about Fedora being mentioned on the site, then it is somewhat like buying the ad place on the KDE site. And while ads may be valid as marketing tool in some cases, it doesn’t feel right to use them as a way to collaborate between FOSS projects.
And if we are talking about the place on the advisory board,
Fedora KDE SIG already does a lot in terms of creating and maintaining KDE userbase. And the way I would like to see Fedora Project sponsoring and investing in KDE project is by bugreports, user support, code contribution, and design discussions.
Relying on money to make the voice of the Fedora Project heard better by the KDE community seems to be unhealthy for both Fedora Project and KDE project.
Our main guiding principle is Upstream First. If we can not convince KDE community to prioritize a certain feature, we shouldn’t use Red Hat’s money to push the community in the direction we want. We follow the upstream and support upstream as a place where decisions happen based on meritocratic principles and not the money investments.
And we haven’t even started with the topic: if we get Fedora representative on KDE advisory board, how this representative will be chosen, what are the guiding principles for them and what is the decision making process associated with this role.
Let’s expand this obviously way too short answer. Disclaimer: I’m a Fedora KDE SIG member, Fedora Kinoite maintainer, KDE e.V. member and Red Hat employee.
I am in favor of Fedora sponsoring the KDE project officially because that would reflect publicly the work that we have already been doing as part of the KDE SIG for some releases now:
We have regular meetings with KDE developers and we discuss and fix issues in collaboration with them.
We follow KDE upstream releases quite closely.
We have several KDE SIG members that are also KDE e.V. members.
We would thus definitely not be “buying our place into the KDE community” as we are already there.
This is already how it works for GNOME with Fedora on two fronts:
Red Hat is on the GNOME Advisory Board
Red Hat funds the GNOME project and employs people to do things it wants in GNOME
In this respect, Red Hat has bought its influence in GNOME. Is it the sole organization doing so? Hardly. But it does. Is it healthy or not? Honestly, I don’t think it’s terribly unhealthy as long as the developers in question are reasonable people that have a collaborative mindset anyway. If they don’t, then it’s a bad relationship.
Being a KDE Patron signifies that we care about KDE and we want to ensure our concerns are factored into the future. We cannot hire anyone, but we can support KDE to hire people to support the project itself. This fundamentally improves the sustainability of KDE, irrespective of everything else.
We are members of the KDE community today and we do work with them on various projects, but one of the problems of just being at just the IC level is that it’s difficult to identify other stakeholders that may be interested in our viewpoint. Strengthening our voice by collaborating with other advisory board members to drive the direction of the KDE project is an important aspect. Every couple of years, the project sets goals to focus the development of the project. This is based on feedback from the board, membership, and developers.
Fundamentally, Fedora is principally an integrator. We do a little development here and there, but we are not able to code up things that we’d like to see in KDE software. However, we can provide the best experience within our goals and parameters and try to make KDE software available in more ways.
If we actually do this, then we will select a representative from the Fedora KDE SIG to be our primary representative in KDE, e.V.
We do have non-technical perspectives as well. There’s really not a lot from a technical perspective that we need to guide with KDE. We’re pretty good on that front already.
From a non-technical perspective, the things we’d want to push for are:
Adjusting the release cycle cadence to ease adoption of the latest releases
Driving a continued focus on simplicity and usability
Increased care about accessibility for broader user bases
Marketing and advocacy for Fedora KDE to the broader KDE community
Expanding OEM relationships and leveraging them to widen the visibility of KDE software
The patronage indicates the opposite: we do not have a connections to KDE but want a connection to the maintainers and developers. So, if @ngompa would take the seat in the advisory board, it would enable him to get contact to @siosm - but @siosm is already part of Fedora himself. Imho, the information about the patronage is very clear about that.
What can KDE do with these resources that the Fedora KDE SIG cannot do here? I am absolutely for providing the Fedora KDE SIG with all resources it needs. And there is no reason why KDE e.V. should not use it as well, which will be no problem since several members are members of both communities. This brings them closer together I think than reducing the connection to a money transfer (finally, we are talking about a donation).
I agree with @bookwar that we have other means, and these might create closer ties. Which brings me back to my original suggestion, why not providing the Fedora KDE SIG with more resources?
Another important point: there are lots of great projects that are linked with Fedora. Why to donate money to KDE but not all the others? I think we should in this respect also stay neutral against the upstream. Everything else would be unfair to some (or cost a lot of money). Although I would be very happy if Red Hat would engage with KDE patronage.
You’re thinking as if I want to talk to developers. I can talk to all the developers I want normally. It’s the non-developers that steward and support the project that are harder to get to.
To put it bluntly, they can hire people and execute on projects. Things like usability studies and expanding partner relationships require money, and aren’t things Fedora can or will do for us. But KDE can do that for us and the broader KDE community.
Red Hat contributes employees and/or money to the vast majority of non-KDE projects that we actually include on the KDE spin and Kinoite. Their coverage is quite vast.
Red Hat would probably never engage as a KDE Patron itself, since they stopped shipping KDE software in RHEL 8. It’d be a really hard sell. But Red Hat supporting Fedora being a KDE Patron is much more likely because Fedora supports KDE for both Fedora Linux and Red Hat Enterprise Linux. In this respect, we’re the real stakeholder.
This will never happen. We’ve tried to do this before and have been rejected already because Workstation already exists. That’s why we’re trying to have the Fedora Mobile Edition use Plasma Mobile, because that puts a KDE solution at Edition status.
As for the “buying influence” aspect, that’s one way to look at it. I wouldn’t personally characterize it that way, but it’s not wrong. To me the motivation isn’t getting a seat on the board, it’s funding an important upstream. That we get a seat on the board is a nice perk.
I’d love it if we could fund every upstream. After all, free software may be free to use, but it’s not free to produce. Unfortunately, we can’t fund every upstream. So our choices are to provide support to none or to provide support to some. I think it’s better if we can support some, even though that makes work for us in figuring out how to make that decision.
That’s supposed to be the conversation now. We need to develop a framework for making this decision and then decide if KDE patronage fits in that framework.
@ngompa In all your examples you refer to Red Hat, which is not Fedora. There is a reason why both are separate. Now we are talking about a community to become sponsor of a community. Another abstraction layer with many implications.
I prefer KDE myself and would like to see Red Hat becoming a patron and to further foster KDE, and I would support any means to encourage Red Hat to do so. But not to replace Red Hat, or another potential patron, I mean there are others, too.
Another important question just came to my mind: can Fedora legally become patron? If the patronage is a type of membership of a German “eingetragener Verein” (transl. registered association), it would need to be a legal entity, which would imply Red Hat. That question should be clarified first because as far as I read the KDE e.V. charter, patronage is a “Förderndes Mitglied”, which means something like sponsoring member: so, a legal entity becomes member of another legal entity. The charter is not absolutely clear if patrons are sponsoring members as the patron status is not defined in the charter on itself.
Has the Council already clarified that?
Makes sense to me to discuss that first, and then of course if the patronage is legally possible.
Good point. I think we may have tricked ourselves into finding a problem to the solution and thus not having a common goal yet.
Becoming a patron of KDE project may mean several things, and serve different purposes. And it is an implementation detail, rather than the goal on its own.
So we should probably start with defining what we are trying to achieve. And then see what is the best solution to those.
For example, if the goal is to simply support projects we use and not to steer them into a specific direction, I would rather consider a different practice: at the end of the budget year distribute whatever budget is left unused by the project itself to some of the FOSS projects based on certain criteria (vote, percentage, survey results, contribution flows,…)
I saw this executed by several FOSS projects and I think it is much less controversial, when support of the upstream happens without conditions and obligations.
Important is to have a transparent institution (framework) that regulates the distribution of funds, so that all potential projects are equally considered without subjective preferences.
I would not worry about this. The details of working out sponsorships to KDE has already been done for last year’s Akademy sponsorship where Fedora (through Red Hat) paid to sponsor the event. This would merely be a recurring one to KDE itself.
I think maybe we should set some ongoing expectations and benchmarks for the sponsorship in addition to the mentioned “rubric” (or as part of it). Both what we’re getting out of it and what we’re putting into the relationship beyond cash.
If we decide that this meets our criteria and we decide to go ahead, I don’t want to re-debate it every year — but I also don’t want it to be set in stone forever. We should make sure it continues to be a good idea.
Since this conversation has stalled, we’re going to mark the ticket as deferred. Not that it’s bad, just that we don’t want to give the false impression that someone is currently driving it forward.
For the next step, we need a draft framework / rubric, with general expectations and guidelines for setting benchmarks, etc. This doesn’t need to come from the council — or from Neal. Anyone interested can help!
