 I'm Matthew Miller, the Fedora Project Leader and this is another Fedora Council video meeting. You may have noticed that I've painted my room here to a slightly lighter blue, which hopefully makes it be a little bit less murder-addicky as Adam Williamson once described it as. I find it a little more cheery, at least, but still coming at you from there. We try not to do all of our work in meetings in the Fedora Council. In fact, when we started out, we tried to not have any meetings at all. It turns out that's a recipe for never getting anything done. And we also found that having a video meeting every now and then, in this case once a month, is a good way to have a high bandwidth conversation that is also something we can share with other people on the YouTubes and social media and everything. So once a month, we try to do a check-in with the part of the project that we're interested in, that has something interesting going on, might need help, or something like that. So this week, this month, we're looking at Fedora Server. We have Peter Boy from the Fedora Server Working Group tell us about that. I'll give a little bit of background first, which is that Fedora, ever since it started in the early 2000s, has always had a lot of server usage, people at universities, people coming in from using it for all sorts of different things. Obviously, your desktop is pretty important to us. People in the media often think of Fedora as a desktop operating system, but somewhere around 30% of the systems out there running Fedora in some way or another are running in a non-desktop context. And so that's always been an important voice. But it started to get to the point around Fedora 20 or so, where there's a lot of conflict between what was needed in the Fedora server operating system, the file system layout and those kind of things, and choices for the desktop. And so the Fedora board at the time, which was the precursor to the council, decided to split into three additions. We had Fedora Server, Fedora Workstation, and Fedora Cloud. And those things have kind of evolved. And that's been pretty successful, and it's allowed the server to make different implementation choices than the desktop, which makes sense because although we share the same Fedora Linux base, there's often things that make different sense for different use cases. But it has happened that while there's kind of a lot of energy around the desktop and things there, the server group sort of trailed off to just one or two people working on it, and there wasn't a lot of excitement and energy around it. And so we were talking about de-additioning it. We've got a bunch of things in Fedora which are important, but which are not yet actually one of those top three additions that we kind of focus things on. So we talked about, you know, not canceling Fedora Server, but moving it back to non-addition status. And then a bunch of people said, hey, no, wait, I actually use this. This is important. And so some people, including Peter here, stepped up to kind of do a revitalization and look at the organizational documents, kind of help that group get put back together and kind of set the future direction for Fedora Server. So that's the background, and I'll turn it over to Peter. Thank you for being here. Oh, thank you very much. Well, I think I would start with my slides here. Sure. But I have to, probably, the technique here. I don't have the chair screen. Oh, it vanished again? Oh, fine. Thank you very much. I'm not sure. We tested this just minutes ago. So there we go. Yep, it's working. Oh, I hope it's working. It's working perfectly. Oh, it's my first page. Do you want questions throughout, or should we hold them to the end? No, no. I think it's better to have the questions as soon as they are coming up. Okay. Can you see our video as well, or should I interrupt you? I have two monitor desktops, so I have the screen on one, and you and me on the other side. Nice setup, then. Okay. One name is Peter Boy. The boy, as in boy or girl. It has nothing to do with English parentage. It's the name of a smaller region here in northern Germany. Well, I'm a scientist at the University of Bremen in Germany. It's a horror. I studied sociology, you know, as a major, but applied mathematics as a kind of, as a minor, is it, as a minor. And applied mathematics was a father, the foster father of computer science, at least in Germany as in Europe. And because of this double qualification, my colleagues considered me as suitable to plan the IT equipment of a network of several highly, highly qualified research institutions at the university, and to manage the IT department practically as a kind of side-drop, because my first job was to do research, but the colleagues were thinking, okay, it's not so much to do with IT management. You can do it as a side-drop. The positive sign was I could do what I wanted. I had never, no one to ask. It was alone my decision. Well, in the early years, I choose IBM AIX for our server infrastructure, and later switched to Linux and Intel machines. And after comparing different distributions, I decided to go with Rated, and then to Fedora as a server machine as well, which I felt was the closest to the quality, the demands, and the reliability. In the IBM world, I was so used to, and I, well, I find it good. Okay. Thanks. I think it is good. So, well, as part of the co-focus on adaption to real work processes, it's a topic of documentation. It was a key function in my work, and I had worked several times with different IBM organizations in the field of documentation, including the ITSO, probably, you know, it's an international technical support organization in Austin. Over the last years, I got increasingly frustrated with the state of documentation in Fedora, specifically the server. Fair. Whenever I use the search engine, I trust to solve a problem. The first hits were always Debian or Ubuntu. Ubuntu, I don't know how to pronounce, followed by generic sites like Stack Overflow and the like. And even if I included Fedora in the search string, I didn't get Fedora, but a slightly modified list, but always Ubuntu, Debian. Well, I have various document... I created a lot of documentation for internal use about Fedora and about, okay, Scientific Genux as a variant of Red Hat for our own purposes, and so I got the idea to rework and expand the documentation to make it generally available. Well, and as Matthew called for the reboot session in December last year, I got the idea, okay, it's something for me, I can do that, and I don't have to set up my own website or so. I could use a Fedora one. Well, over the time, the limited focus into documentation evolved in something wider or broader, commitment over time. And when I reflect this development as a sociologist, I myself have to a certain extent become a victim, so to say, a victim of a successful so-called community-building process. Is that the official sociology language of a victim? No, that's not sociology. Well, so I want to avoid to held a sociological lecture, of course. I'm interested in that too, but yes, go ahead. But I think I will pick up some sociological terms and some sociological knowledge to reflect all about the things we are doing, specifically which I am doing. So, well, I use knowledge about community-building to try to assess and to think about what went on during the last quarter of a year to reboot the server working group. Well, the result is, in some cases, an odd mix of strong capabilities on the one side and several minor but accumulating surprising weaknesses on the other side. And maybe, or I hope, we can get some ideas how to improve the situation. Well, if you take a look at the numbers, it's always a fine idea. The December session started with 18 new participants. And if you look at the... Do you see my mouse pointer here as well? Yeah, I do. It probably won't show up very well on the unrecorded video, but I do see it. Okay. But you see a testicle form of such a curve. You start with a lot of people, 18, and the number diminishes over the time. Well, today it's between 5 and 10. So that is one of the well-known challenges of a community-building process. We have... Well, participation decreases over time. That's one item. And the other is it is relatively easy to start something new. You start as a working group, and it is difficult to make it last, to make it permanent. And the well-known challenge in community-building is to permanently attract people and to offer opportunities to participate. That's the second item. So one of the challenges of the Fedora project, specifically the server working group, is to attract people. Well, to attract people, so... Well, if you... Wait a moment, so... The challenge is permanently attract people, and the way to do that is, usually, the project's website. And if you have a look at the Fedora landing page, well, you see the base address is really our base, the permanent base address, fedoraproject.org. It's redirected to a download page. That page is basically about download only, nothing else. Fedora presents unintentionally, I suppose, a mind like, just take it, nothing else to care about. Yeah, actually, that was a somewhat intentional decision when we did that redesign, which was... That was around Fedora 20 as well, so that's seven years ago now. And it really was kind of focused on a user audience brochure. Like, this is for the people who just... Like, they don't want to be hassled into trying to have to contribute. They just want to get started, and then we'll sweet-talk them into it into something more later, perhaps. But I think that's definitely something to look at as we redesign. So, yeah, thanks. I appreciate that suggestion. Yes, I think so. Fedora just amplifies what open source doesn't stand for. In Germany, it is mitnahme mentalität. It's something in English, grab all you can mentality, so to speak. And in the end, it is a kind of misuse of public benefits. So we are, as a project, as Fedora project, we have to... We have... We must have people who engage themselves, who are committing themselves, and we have to create opportunities to do so. And I think the most... Yes, the most important community is... Opportunity is the landing page. I think it's... I don't know the numbers, but I suppose it's a page which is mostly addressed. It's just super interesting, because I was just having the same conversation with the workstation working group last week about kind of a lack of a landing page that kind of integrates both the product, the thing being made, and also the community around it, the working group and the team and all of that. So, yeah, this is really good stuff. I hope so, yes. If you have a look at the landing page, it is here at the moment. That's our famous landing page. We miss the opportunity to promote events that may be appealing for new engagement, you know? We have to excite people to get involved in our projects, and we are missing or we are leaving off any opportunity to do so. Yeah, it is very much a static page. It's very much a static page. That's my second issue here. There is no interaction. There is no way to click anywhere and say, hi, I want to participate. Where can I do that? And the landing page has no information about the project itself. So, sociological terms. Fidova is a site of idiosyncratic community. It's a community with a special, you have to know or you have to acquire special knowledge before you can do anything. It's something like a catholic church in European, you know? You must be initiated. That is, you must have acquired a lot of internal Fedora-specific knowledge before you can do anything. Part of what happened here is that along with the plan to have this brochure site on GetFedora.org, we were going to have something called Hubs that was basically a Fedora social media engine that was going to be on fedora-project.org. But that project ended up being too ambitious for the resources we had to deal with it and it ended up having to get scrapped as sometimes software projects do. And then we never had a good solution to that. So I think we need to go back to the drawing board with the next website redesign and look hard at our resources. We have to actually do things and then tie together some of the things we already have as well. Yeah, that's it. I think Fedora already has a lot of things. EG, I come to the next slide. Excellent. The famous site, Eps Fedora.org. It's an excellent overview about I think perhaps it's excellent. It's an excellent overview about the Fedora structure and where it is done, what and so on. This site was a link to this page. It belongs on the front page and the landing page, of course. And it belongs, let me say, to each footer of any page. So we can maybe get an overview of what is going on where. So we already have a lot as a Fedora project, but so to speak, the Fedora project is very good in hiding all the treasures it has. So... Yeah, Neil makes a comment in the chat that we actually had another thing called FedMenu, which was a little footer icon that did something like that. Yes, well... Yeah, anyways. Can I have another look at the Fedora docks project? If you see the Fedora Council, learn about the Fedora project governance. It's quite inspirational. It's redundant in a way. And it reminds me of a lot of comments I found in redhead programs, such as class hide me and command this, this is class hide me. So there's no additional information, no inspiration. We need something else. There must be a sentence. What is the essence of Fedora Council? All right, that's fair. We must attempt to click to get new information, interesting information. Well, I think there's a much, many is already there, but we have to do some fine tuning or especially physically we have to improve that what we already have. Well, back to server group. Yeah, yeah. This is great though. I find it really interesting. Well, server group is nice or a working group is nice, but the main issue is we have to create items. We have to create a program. We have to create packages. In the case of server group, at the moment it is we have to create documentation. Well, we started with 18 new participants and with a state of yesterday, we have four item creators. We have four participants to participate on discourse to give ideas and so on. We have three lurkers of the initial 18 and we have lost seven. This is never showed up. It's not bad, but it shows participation is a real problem. You have to calculate that you lose a lot of initial interested participants and that emphasizes the need for constantly trying to attract people to replace those which drained off. At the moment we have to... Oh, sorry. I don't have a back button here. Oh, I have a sign. We have 22 active new participants, not the members of the previous group. They are still active, of course, or fortunately. We have now five items creators, seven participants on discourse and ten lurkers. This is a specific structure, I suppose, in groups like, I think, or in community-built groups. You have always a core of highly engaged, highly committed people and you have a larger group of people around which are less active and less committed. The issue is... The problem is you have two issues. Of course, people who are participating on a discourse or lurkers are important. They are indeed, because they are providing, probably providing new ideas, some critique. It's very important to have them. Nevertheless, you have to stabilize the core. You must find, in case of the server group, the five people to organize their work. Fedora used to have some kind of differentiation. It was a service, it was a special interest group of the one side and the working group on the other. In case of Fedora Server, we don't have such a differentiation at the moment. I would like to have such a differentiation back because it's very difficult to decide whom to address to do some work and so on. It's a rather critical issue at the moment. This also parallels the conversation in the workstation working group. I wonder if we have a nest with Fedora session to get everybody from the different working groups together to talk about the sociology of this and how to arrange it ideally. Do you have, in your mind, ideal targets for these numbers? Five, seven, ten. It seems like it's a little lower than it would be nice to be. Is it ten, twenty, a hundred? No, no, no. I know literature. According to literature, you need about seven to ten people as a core. If you have more than ten, the communication gets complicated. Less than five is turning down. That's the summary. The community building working is with a lot of loss and the numbers emphasize the importance of ongoing recruitment process. So the landing page is a very important topic. But the question arises in how to give a boost to the item creators. The general answer is we need awards, back incentives for those. Again, Fedora has some of those in place, badges as a form, a way of incentives. At the same time, it misses various opportunities, for example, in the landing page. An example I can speak about is documentation because documentation is one of the biggest working projects at the moment for the server working group. Well, G's landing page doesn't mention any, there's no mention of documentation. There's a small button at the top which says help, but no documentation at all. The same is on the download pages for the server and the other editions. No mention about documentation. So to say Fedora is pretty good at hiding, it's treasure, as I said. We have documentation, but we don't announce it. And it is an issue with those who are authoring documentation. Visibility and public perception is one of the important incentives for writers. Mind you, an author generally writes a so-called publication. It is from public. It's a pretty good pamphlet for a small circle of illustrious or Fedora experts who already know everything anyway. So the first step for improvement, I think, is pretty obvious. Use the popularity of the download page for documentation. It's a pretty simple search engine optimization. Oh, it's a quicker word. And the current state of cycle current stage is a kind of vicious cycle, Teufelskreis in German. Lack of visibility and documentation leads to perception of difficulties. Devil circle in German? Yeah, okay, yeah. But it affects dissemination of Fedora, which in turn leads to low commitment to create or maintain documentation, which in turn affects dissemination and so on. Unfortunately, from my perspective, in Debian or Objunt to this circle works the other way around because they have a better presence of documentation. If you read about Fedora, you often find the characterization technically excellent, a work of engineers for engineers and implicitly not for non-engineers. So that's a kind of presentation. Yeah, so I think that's exactly what the idea of having to get Fedora page, not have technical contributor stuff mixed in with it, was sort of trying to address that in some ways. But I think we can see there's some things we've learned in seven years about how we could do that better for sure. Right. The work on documentation I think is if I would have a better public presentation of documentation, it would make my work on Fedora server documentation a lot easier anyway. And I think it's the same for all the documentation for Fedora. At least my impression is you should make a reboot session for the documentation group probably as soon as possible because it's quite dormant, I suppose, and there's a lot to do. Yeah, thank you. And as I mentioned, but I don't want to speak about it so long, the participants are one side. The other side is what the participants commit to. So the subject of the commitments that is Fedora project as an organization. Just short here, I think Fedora's profile is beginning to blur. We had ten years ago, five years ago, we had a Fedora project. Everything was Fedora project. We had a Fedora hosted Fedora project. Now we have Fedora project. Yes, we have Pagore. I don't know the pronunciation. No one does. We have Github. Fedora is all over the world. But it misses a sharp profile. So it loses some of the grip. Some of the profile where you can commit to. So you don't commit to Fedora project anymore, but you commit to some project at Github or something like that. And that's not a philosophical issue. I don't know if I have the other moment. Oh, it's the wrong... Oh, nice to be... Oh, sorry. Okay. I'm missing some slides here. You can see the problem, for instance, with the cloud group. If you are looking for... if you pick up your search engine and looking for Fedora cloud documentation. The first hit is Fedora cloud.reads.dox.io. And there is a documentation which is from 2016, I suppose. And the second hit is a Github address. Fedora cloud at Github. There is the original source code for the documentation. And then cloud.google.com for the next hit, yeah. So I think it's... But the problem is, we present Fedora as a... if outdated, nobody cares about, it's difficult to use, better use anything else. So I think we have to... in the long run, at least, we have to concentrate and... back to concentrate on Fedora. It's a Fedora project address. But that's a short excursion because I think that's a bigger problem. The discussion we had, I just think three years ago, or five, I don't know, about how important is Fedora infrastructure. Okay, then are my ideas so far? Well, I'm ready for questions. I really appreciate this, and I really like the sociological angle you've taken here. I think it's a lot to think about. And it really is well-timed with what I was talking to the workstation working group. And the cloud working group, and actually CoreOS, for that matter. All these people kind of have the same kind of problems. So I think it does seem like a website redesign that takes into account these paths and the addition product to team, tie, getting people to documentation, and having a little more excitement. All of those things seem good. Marie was not feeling well today. I think some of the websites and mindshare redesign is kind of in her wheelhouse. I'd like to hear her thoughts on that, but it seems like an initiative around that is probably something we should focus on in the next year. Right, Neil Gampas says in the comments, Fedora CoreOS is off in their own corner. They don't really want to be off in their own corner. So we can kind of try and pull that in, especially as I think they're having a lot of user success as well. That kind of ties into the whole, let's make this one Fedora project thing rather than things off in different corners. Yeah, so that's one thing. And then the documentation in specific, like the documentation site, the front page for that kind of grew up. That wasn't really well designed. It just sort of is organic as bubbles started growing on the page with each little thing. And that could use some more thought as well. I guess I do have one question about the pathway, which is the Fedora join group is kind of where we steer people who are interested but aren't quite sure what they're wanting to do in Fedora. Have you had interactions with that team at all and are people coming to Fedora server from there? Is that something we could improve? It could be. I don't know about it. You know, I'm half initiated. All right. I was happy to find a way to put an article to Fedora magazine because it's another or was another issue. I expected to create an article and suddenly I was on a page to have to create a project because a project for me is something different than an article. But in Fedora magazine, it's the same. But there's a lot of small pieces. And so I didn't know about join. But if you are going to resign the web pages, I would be happy to join that. Okay. Yeah. I'm thinking maybe we'll have maybe a preliminary meeting about this with some of the people from all the different working groups and different interests. I think that would be a useful thing to do. Neil also says in the comments here that it would be nice to have the spins, the non-edition things that people work on, KDE, obviously, I3, XFCE, and also design suite and things like that. Those are kind of hard to find in the current design. And I definitely agree about that as well. Those are important parts of the project. And that design, again, came apart with the best intentions, which was that previously we had a lot of user feedback that when you went to the page to get it, you had 500 choices to make and that when you just wanted to get started, you didn't necessarily know what all these things were. And so it was kind of overwhelming. So we wanted to make it less overwhelming. Yes, and Neil also says ARM and other architectures. Fedora for mainframes for the few people who do actually use Fedora Server on a mainframe. Making these things easier to find is part of that as well. Do other people have questions? Neil says he doesn't know where to find 64-bit ARM or PowerPC stuff. I do. At least, yes, it's alt.fedoraproject.org. You can find all of those things. But, yeah, that's like another secret. Anyone else have comments? If not, this has been, like I said, really helpful and I think I do have some next things to think about and all of us in the Council have some next steps for things we want to support. Peter, are there other things we can do? Yeah, I have a question. You're talking about awards and incentives and things. Are there things we can do to help grow the team, incentives and rewards we can give? We actually do have some budget to spend on things. In non-COVID times, we might talk about doing an in-person meetup to get the team together. We do have travel budget and stuff when travel is again possible. In COVID times, what are some things we can do? Traveling would be nice, of course, because it's quite, or to discuss complex issues. The IRC is quite, well, it's problematic. So, I don't know if you have the option to make a video session and to discuss it the way we are doing it here, because I think it's quite costly, I suppose. I don't know. This video doesn't cost Fedora anything Gret has paying for it. So, it comes out of a different budget, not our problem. And we also actually, as part of our new matrix server coming online, that's going to be moving IRC to a matrix element instead. That actually also will have a JIT-C video server that's part of it, and that'll be hosted. So, that'll be easily available to the group. So, that hopefully will make those things better. We can also do things like team t-shirts, things like that, other swag for people who are active in the group. If that motivates people, that's something we can provide, you know, team hoodie. That's a thing. A lot of people like that. And if there are other things you can think of that help, you know, reward people for their involvement, let us know. We can help provide those things. I have to think about it. At the moment I have more basic issues. I have to get people to write texts or to read files. We can also, if there are badges, like if you'd like a wrote Fedora server documentation badge created, you can ask the badges team to make those specifically for people. So, that's something that might be useful to you. That might be useful, yes. It might be a start, yes. Yeah. Basically, you showed up to a Fedora server meeting. You can make all these different badges and give them to people. That kind of does two things. It gives motivation, you know, people who like to get the badges can do that. And then also people who are just kind of looking, they can see, oh, here's the thing I could do. Oh, I could, you know, looking through the options. But I think the most important thing is to give feedback. And there are various forms of feedback. The badges is one of those. And I think it's a good idea. Another thing, you know, I have to learn about badges. I don't know anything about it. But I think I would find it. Another thing we could do, Shimantro does a Heroes of Fedora QA every release and kind of has a blog post about all the people who did, you know, good things. And then we could have a Heroes of Fedora server post occasionally that kind of gives people credit for what they worked on. That's my ideas. I think there are more options as I thought about, yes. Yeah. All right. Good. I'm glad to hear we can do that. Anybody else have anything? Thank you for doing this, Peter. Oh, all right. Thank you very much. Yeah. I personally am really happy to see that Fedora server is slowly but surely coming back to life. And, you know, wherever I personally can try to, like, be part of it, I will. Hopefully it'll be more successful than the first go around. But, you know, I'm glad that you're trying and I look forward to seeing, you know, how things move forward with Fedora server. Oh, I'm quite optimistic because unfortunately, I'm not alone. We are a group of about five to six now and we are quite active as a thing. We are on a good way. That's really exciting. We have some more comments in the chat. Grayson says that he is able to help you with getting the badges. Oh, that's good. That'll be good. And Vipple gives the link to the issue tracker. It's paggar.io, Fedora badges. Ben, are you still around? What do we have planned for next month? Ben has died and fallen in the corner. Ben had children to deal with, so that may have happened. I assume he's still alive. So I don't know offhand what we're doing next month, but we will do this again next month with something else. Again, Peter, thank you so much for this. This was a great informative, useful session, and I hope we can be helpful to you as well. And again, yep, thank you for all your work on this. Bye, everybody. See you next time. The next, the next is Fedora Linux, the next digital public good by Justin. Oh, yes. Justin, digital public goods is a thing we're going to be talking about. So that will be, that will be super interesting. It's kind of about this, you know, the comments and things that are good, you know, not just a tech project, but something that's good for all of humanity. So, yeah, I'm looking forward to that. I'll see you all, well, online in the meantime and here again next month. Bye, all. Bye-bye.