 Great, I'm Christian Glumbick. I am an associate software engineer at Red Hat. I joined one year ago. And I'm also the OCD working group engineering chair of the engineering liaison for Red Hat engineering. Maybe we'll start off, well, first I have to say, I think I've never done a presentation in front of that many people. So excuse me if I'm a bit jittery. I'll start this off with a little explanation of what OCD actually means. Does anybody know? OK, Diane, may I tell? Yeah, so I think it's the OpenShift community distribution part by Kubernetes officially, but it's not an acronym. So OK, Diane is probably the best we've got so far. All right, let's get going. The agenda today will be OCD4, the roadmap. Then I'll tell a little bit about the current status where we're at right now. And then a little introduction to the OCD working group. So OCD4, roadmap. We developed a few guidelines that would always have to be valid for OCD releases. Everything on the roadmap is still not formally agreed upon and may be subject to change, but these are the things we thought internally and also with some community members would make sense to always have be true. So that would be we use Ignition version 3 or the spec version 3, which currently OCP runs on spec 2. So with OCD, we'll start from spec 3 right away. And then we'll always support Fcause as a bootstrap and no-toast operating system. So we can always say OCD will run on Fcause. That doesn't have to be exclusively, but it's one system that we always want to support with OCD. Fcause is Fedora-CoroS. So the OCP platform is R-Cos, Redhead-CoroS, or Rel-Cos. There's different names for it. And as it's not a standalone product, it's not really a thing. But in Fedora world, we have the Fedora-CoroS, which is sort of the Fedora atomic successor. Yeah. So we want to avoid unnecessary divergence in OCD from OCP, at least for our initial release. Stay really close to OCP and align and standardize APIs along the way where it's possible. And all the platforms that are supported by the union of R-Cos and F-Cos should also be supported by OCD. OK, then we have divided the roadmap into three phases, starting with phase zero, which is where we're at right now. And the goal of that phase is to enable the community to have something to test and to actually use and tinker with. So that's where we're at right now. I'll come back to this in a bit. In the midterm, which is sort of the end of that term will be the first general release, which I hope we can make happen in the first quarter of next year. No promise, but I think we're on a good way to get there. So in phase one, we'll just do a little bit of hardening, more CI testing, adding things to make sure that our release is actually working the way it should be. And then in phase two, that's actually the most interesting phase, or it's going to be the most interesting phase, I think. So that's after we've released OCD. We really want to make sure the community can contribute. So we've been thinking about ways how to do that we have a community repo on OpenShift Community on GitHub. We have an enhancement repo where you can add design proposals. So all of these are sort of to facilitate contributions from the community. And the goal behind this in the end is to make OCD a superset of functionality that flows into OCP. And what we want to do really is we want to use the same code. So we don't want to branch and diverge too much from OCP. We just want to build it on top of Fedora for the initial release. And later, I'll come to that. There may be subworking groups that will enable more OSs. So we've definitely thought of that and want to facilitate that and help enable that. If you want to have a look at the roadmap, it is, did I put the source? Yeah, there's a little source thing down here. I'm sure you can't read it. So right now, it's a draft. And we're going to propose or do a call for agreement in the working group. So I'll link to the working group later. You can find the roadmap draft with more detail in there. And we'll publish that on the mailing list. So the current status where I would write now, we're in phase 0, which is a little bit, it's hard for the community to contribute at this time because internally, we have to make things ready for that. And that's mostly stuff Reddit people do anyways. So we are probably the best people to do that. And we want to do that quickly as well. So right now, we're setting up CI for RPM builds of OC and HyperCube. Those are dependencies on the ARCOS side, on the OS side of OCP. And we want to package them continually with the CI system for Fedora. So we can use them in the Fedora CoroS Composes. And then the second thing that's going to be unblocked very, very soon, probably next week, is when we switch to 4.3 master branches, we will be starting to work on the readying of the installer and the operators for IgnitionSpec3, which is especially the installer and the machine config operator will have to be. Yeah, they have to be ready for spec3, which they aren't yet. So there's some changes in there coming. And after that, when those two are already will be in phase one, which is all the things should be ready and we'll do some hardening before the release. Right. The next point on the agenda is I'd like to very shortly introduce the OKD working group. I'm not sure. I wasn't here in the morning to probably introduce the working group a little bit. I thought I'd just do that as well from my point. So right now, we formally came together the first time. Well, we agreed upon a charter, which is our governing document last week. And so we're formally a working group now. Diane is the community chair. Then we have Danny Komea, who is our external to Red Hat engineering chair. And I am the Red Hat engineering chair on the working group. We have the repository. So if you're interested in contributing or just staying up to date, I'd suggest you follow that repo because all the agreements will be made as issues or PRs on that repo. We also have Google Group, which serves as our mailing list. We'll make sure to always send out the information on the other mailing lists, the OpenShift users and OpenShift Dev mailing list as well. But this one is really a bit more focused on the OKD4 effort right now. So you're also invited to join that group. And then we have a Slack channel as well, which is the one you may know already, OpenShift Dev on the Kubernetes Slack. We also have a bi-weekly meeting, which you're invited to join and say hello to as well. I think we don't have a calendar yet, but we'll set up one real soon. So you can subscribe to that, and you're definitely welcome to say hello on the meeting. This is my short introduction of OKD4. If you have any questions, I think I have some time left. So if you have questions, I'm happy to answer. How many of you in the room? So OKD and Origin have been around for a long time. But with the 4.0 upgrade to the operators, there's a big shift in the way that we're deploying OpenShift and OKD. And that has really impacted on the delivery of OKD4 because we have to refactor it to run on Fcause or Fedora CoreOS. So how many of you? There were a few of you in the room that were OKD users. There was a couple. There you go. All right, all of you good folks. We love you all, and we'd really love to have you take a look at the charter, not the charter, but the roadmap and help us make sure that we're on the right path. And if you have extra engineering cycles, we'd love to have as soon as we get these first RPMs out and this first delivery of phase 0, we'll be sending an announcement out to the OpenShift Commons list, the Dev list, the users list, just so we can get as many of you to test and play with it as possible. That is our really big ask because it's your feedback that's going to really help us make this happen. And Fedora CoreOS has got some timeline plans for being GA in Q1 as well. But I think we'll probably have our release out earlier. So that'll be key. Yeah, we decided to not depend on Fedora CoreOS. There will be a version of Fedora CoreOS that we can work with. So this is another one of those at the beginning. You missed it, my jellyfish diagram. The interdependencies of open source projects really come into play in coordinating. So we have people from Red Hat who are engineers working on the Fedora CoreOS effort, on the OKD, on the operators. We have people in ignition. And a lot of those people are the same people. But anyways, we would love to have your feedback on that. So stay tuned. And if you can, the next bi-weekly one is on October 15th. Right, let's get one because. Yeah, we have some travel issues. Somebody's going on PTO. And they usually are around 1,700 UTC. We may shipped it a little bit earlier to do that. So we're trying really to accommodate as many of you as possible to come in and participate and give us feedback on this because it is one of the biggest initiatives that we've had on the open source side that really needs as much community support as possible. So we're really happy. There's three to 400 of you in the room right here today. And all of you who raised your hands for OKD, we love you. And we love you to help us out in this next push to get it out there. And yes, it does mean OK, Diane. So you've all just agreed to help. So thank you. Let me add to that. Once we have the release out, the OKD working group is really going to be, I hope, is really going to be a group, a platform to really be on the edge of development, to really test out new cool stuff and get that into OKD first and actually have it testable. So yeah, I think there's lots of interesting stuff coming, like supporting more platforms, not only clouds, but also maybe IoT, more use cases. There's so many things that we're going to look at. I think it's going to be really, really cool. And I'm looking forward to your contributions. So thank you very much. The other thing that I want to just keep holding up the applause, the other thing that I'm going to mention is because we have five minutes before lunch is ready. So I'm just going to keep doing this. Walk slowly to the lunchroom. The code ready containers, we're going to make sure that we have a version of that that supports OKD4 when we come up with a release. So if you start playing with the OKD containers now, you'll be ready to work with that on OKD4. I'm not sure we can make the code ready containers release for the initial release, but it's going to come at some stage. With the code ready containers, there's just some things with the libvert support and also being just one node, just one node cluster that isn't really supported too well by the operators right now. But we will get there, I'm sure. The other thing is everything that we do in the OKD working group, again, I mentioned it earlier, we record everything's open. You can join the OKD working group on Google Groups. It's an open community. All of the video, every meeting is recorded. So if you're not in the right time zone or you want to watch this in the privacy of your own homes on the weekends because it's so exciting, you can go to RH OpenShift on YouTube and look for the OKD working group playlist. So some of the conversations are really bleeding edge and really interesting, at least to me. And it's like stuff with the code ready and the libvert conversations. And we'll be having people from the Metal Cube and the code ready team coming and talking at the next one about how to make these things all happen. Meet new people.