 Thank you everyone for joining us for the working group meeting for July 20 of the year 2021 and I will pop the agenda into the chat for folks that don't have it. And there we are. So let's do a quick agenda review real quick before we get started. Does anyone have anything that they want to add? Is there anything missing or anything that you'd like to change? All right. Sounds like folks are good with what we have. So let's start out with introductions. I always like to do this because there's always going to be new people watching these videos that are getting put up on YouTube. So we'll start out. I'll just go across my screen. Brian, go ahead. Hey, I'm Brian Innes. I work for IBM, but I'm here as a hobbyist. I run OKD on overt and got a homelab set up. All right, Mike Rocheford. How's it going? Go ahead and introduce yourself for the folks that might be. I was like in the middle of doing other things. Hey everyone. I'm Mike. I am new to Red Hat as a junior solutions architect. Often the Northeast, Northeast column sales division, part of the acquire pod. Been hanging around the OKD working group for starting around like, I don't know, like March last year. I'm just kind of in a more of a lurker swatch. What's going on? But I would like to take part in some of the container six stuff moving down the line if I have the time to do so. Excellent. Phillip. Hi, I'm Phillip. I'm working for a for M. And currently I'm regretting the OKD stuff from 311 to 47. So I have some problems, but yeah. I hope they will be solved soon. Thank you for joining us and three team. Hey everyone. So this is the first time that I'm joining this OKD meeting. I'm from Red Hat and I'm looking to contribute to OKD. Yeah, that's all. Thanks for joining us, Neil. Hey, y'all names Neil. I work at NATO as a senior DevOps engineer, but I'm here representing the four community to help make sure that OKD and the and the open shift teams are able to communicate effectively with Fedora teams and get the good stuff in there. Basically easily and quickly. And I'm also here kind of as a hobbyist to, to, to like. Have OKD working for the hobbyist type folks. Because that that matters quite a lot to me. Mike, oh, Mika. Hey, everybody, I'm Mike McKeon. I'm an engineer at Red Hat where I mainly work on open shift cloud platform or container platform and cloud infrastructure team. I just love OKD. I think it's a great community and whatnot. So yeah, just here, open out. Excellent Bruce. Yeah, hi there. I'm Bruce. I'm an instructor at BCIT, which is a technical school up. North of the US border and frozen land called Canada. And, you know, so we're sort of on on break. So I've been distracted with a whole bunch of other stuff so far over the summer. But I'm going to have to get my network up and running with hopefully 4.8 going into the fall if that happens. Excellent Christian. You're muted my friend. No, he's running for 4.8. Try now. No, I'm seeing the icon go on and off. That's weird. Does it work now? Can you hear me? There we go. Yep. There's audio. Hello everybody. I'm Christian Glomick. I'm a software engineer at Red Hat. I've also been co-leading the O.K.D. working group, although I've set back a little bit to focus on more internal work at the moment, which will circle back to O.K.D. I'm currently working on the ARM enablement effort. And yeah, so hello everybody. Excellent. Welcome. And we have Chuba, whoever that may be. You're on mute if you're speaking. Well, we'll move on until if you decide you want to say hi. Timothy. Hey, hi. I'm Timo Teravier and I work in the Chorus team at Red Hat. So essentially, you do a lot of the work around Fedora Chorus, which is the basis for the OS at the base of O.K.D. Mohamed Reza. Hi guys. Sorry, tonight I changed my laptop and don't have webcam, so you don't have my screen. I am actually a newbie in the cloud community. It's about a month or two. I work, I'm coming to these meetings. And this week I am a Bachelor of Science in Electrical Engineering. And finally, finish my Bachelor and happy to be here. Excellent. Congratulations, Amy. Thanks. Hi, I'm Amy Marish. I'm a Principal Technical Marketing Manager at Red Hat specializing in OpenStack. But I'm very big in open source communities and helping to manage them. And that's why I'm here to help out. Thank you for joining us. Shree. Hi everybody. I'm Shree. I work with Neil Adato. And we use Kubernetes and Red Hat at work and OpenShift and everything. But I'm here, I think, along with several people as part of the sort of hobbyist sort of homelab community trying to get O.K.D. more traction in that space. Excellent. And Fabian. Hey there. I'm also an engineer at Red Hat. Working on the Operator SDK team. And so I like to kind of watch these meetings from that perspective. And then I'm also a hobbyist in the homelab space still running OpenShift 3 in my home cluster. But definitely excited to upgrade to OpenShift 4. So following these discussions for that purpose as well. Excellent. And Brian, did you go yet? I don't think you did. I started. Oh, that's right. You started. So we've already made it all the way around. Yeah. All right. Excellent. Oh, and I'm Jamie McGarra with the University of Michigan co-chair of the working group and University of Michigan. We're using O.K.D. for in a variety of places within the university, the one that I'm particularly involved in at the moment is with a consortium called ICPSR of many different universities and organizations doing social science research. So a lot of data and number and survey crunching and whatnot. Let's jump now to the next item on our agenda. And that is the release updates with Christian. What do you have for us? There's two things I think that are noteworthy. Vadim has been looking at preparing O.K.D. 4.8 and 4.9 nightlies. So that is probably assumed to land if it hasn't already. And very, very importantly, I think we have finally merged the F-COS and master branches of the installer. So the installer for O.K.D. can now be built off of Mainline, which I think, yeah, this has been a long time in the making, and this is a great milestone for us. Unfortunately, the MCO, which we had merged at some point now is still a forked again, but that's only, I think, five comments. And we're definitely working on getting that merged back in again as well. But, yeah, getting closer and closer. So that, I think, yeah, that's a really big, important milestone for us internally as well for just making O.K.D. one important part of the main branch. So that is, in the installer, that has happened now, finally. And I think, yeah, Vadim had sent me some things, yeah, O.K.D. 4.8 and 4.9 nightlies preparation. I don't think there's anything else right now. Okay, thank you for that updates. And thank you, Christian and Vadim and all of the people that have been doing the work to get the installer in the position that it's in and to get O.K.D. In the position that it's in right now. I very much appreciate all of that. And by the way, Christian pointed out, and actually it was Vadim, I think that pointed out that it was just a year ago that O.K.D. v4 was released. So it's just over a year that that O.K.D. has been a thing for version 4. And I think a lot of things shifted at that time. And a lot of great things have come out of it. Actually, let me jump right here and jump in here. And one more thing that I think is kind of, it's not strictly related to O.K.D. yet, but it's definitely release related. So for OpenShift, we have just released a developer preview for ARM 64 architecture, which you can download. I think you'll need the developer subscription for that. But there is a blog post about that. I've linked it here. I'll put it in the notes as well. So if you want to try out OpenShift on ARM, you can do that right now. We will hopefully soon also get O.K.D. on ARM. Timothy might be having some updates there with regards to the Fedora CoreOS AMI that is still missing to run the AR64 image on AWS. But yeah, that is definitely on the horizon for us as well. So we're working on that. And if you can't wait, then you can already install OpenShift as a dev preview there. Excellent. Very, very cool. All right, let's move now to the FCOS updates. So go ahead and take it away, Timothy. Sure. Thanks. Right. So as I've been mentioned, I'll talk about ARM 64 support first. And the work is still in progress. I think this deposted one AMI that you can use for testing, but it's probably available only on one region if I remember correctly. So it's definitely a test AMI that will go away at some point. So don't rely on that for anything serious. I cannot find a link for now for this one, but this should be somewhere in the Fedora CoreOS tracker. And yeah, so ARM 64 support is coming along and should be available at some point. Then after that, I don't have anything specific new. Just want to point again to things from last meeting. The first one, we're still looking at Fedora certified changes as they come along. So this is like a continuous process. Make sure we don't miss anything. We're still having discussions around how we should and wouldn't manage. Also on Fedora CoreOS. So this does not exactly directly impacts OKD because you ship with your defaults prebuilt in. So, but well, maybe some inputs, if you have any inputs on that, it's still welcome off on the Fedora CoreOS side. And finally, we had last and last meeting, we had platform support discussion. So we were not of time. So if we still have any questions around platform support for Fedora CoreOS, feel free to read them today so that we could discuss that. Summary is still in the HACMD for which platform is supported for Fedora CoreOS. But if anything, if any other questions are raised, feel free to ask. Is there anything, anyone have any questions right now for Timothy? Go ahead. Yeah, is there anything we can do to help out with that? You guys are targeting AWS. Is there going to be a parallel effort? Is there going to be a home lab hat on for like raspberry pies or similar? At the same time, or are you? Or anything that is like affordable bare metal type stuff, like it doesn't necessarily have to be raspberry pies, but like something that is sub, sub $200 per unit to set up an arm based cluster. Right now, I think the biggest challenge there is the memory that's available on those devices. We can't install OKD or run OKD with eight gigabytes and even 16 is probably not enough. So until there are devices with 32 gigabytes available, we, it won't be easy to say the least, might not even be possible at all. But once we have those 32 gigabytes of RAM available, that should really be, it should be doable then. Yeah, and if you can, I don't know, solder your own memory onto a Raspberry Pi far, then you might be able to do it early. But yeah, that would be a really cool one post. Let's go and talk to gearling. I'm sure he's done it. Can we back up for a second? Let me, for people that might be watching this that aren't familiar with the excitement around arm and why the excitement. Let's actually provide a little bit of context in terms of why arm as an architecture is so exciting to run clusters or run anything on. In short, right, better efficiency fraction of the power utilized for the power that for processing power that you get out of it. If someone wants to be more specific, go ahead and jump in there. I think yeah, that is the biggest factor here. Just cost saving and also more energy efficient. There's some more availability of hardware at the lower end. It's way easier to find a $30 arm board than it is to find a $30 x86 board. Obviously, OpenShift will, or OKD, but then also will only run on SPSA certified hardware or comfortable hardware, which is the arm server ready certification. Which means in order to install Fedora on it, Fedora CoreOS, or OpenShift REL CoreOS, you will have to have that compatibility, which essentially means that the devices are UEFI capable or ACPI capable of booting a UEFI image without having to go through the device tree boot process with Uboot. And that is actually available for Raspberry Pi 4s. There is the UEFI firmware for RPIs, which is RPI4-UEFI.dev, I'll put it in the chat here as well. You can run the OKD installer and use it that way. The only problem there, as I mentioned, is that Raspberry Pi's don't have enough RAM yet. But yeah, essentially it should work. That is still super exciting. Just having ARM out there opens up a lot of possibilities. Especially not even just for your home lab stuff, but I'm thinking more like edge deployments and things like that as well. Really, really cool stuff. There's something that I haven't heard mentioned that I've seen in other places though too, is that like on the other end of the scale, there's a lot of work being done in the ARM server performance kind of sector. And I've seen predictions that in the next year or two, we're going to see a huge disruption on the server and from people running very high performance ARM devices. You know, this is like on the back of the work that like Apple is doing and whatnot, where they're showing market performance increases with the silicon they're building. So like, I don't know how true that is, but if that's true in the next year or two years, we could start seeing a huge presence for ARM devices in the cloud as well. Absolutely. I think that is really the case right now. There's only two or three devices that actually have this SPSA certification. And we're really only at the very beginning of this. And, you know, the more powerful those devices become, the cheaper they become. So maybe with the Raspberry Pi 5, when it comes out next year, or if it comes out next year, maybe that'll already do the trick. And, you know, so really, I think there will definitely be much, much more, you know, many, many more hardware parts will be available. And right now, those are still really expensive, those servers, because, you know, these ampere, whatever, they are really expensive devices at the moment. But once this really gets into the consumer market and is produced in mass, prices will go down and this will become available for the hobbyists among us to run their own OKD cluster. That's what I strongly believe. I'm not super tuned into the ARM ecosystem and what I've, the stuff that I've done. But has anybody, has anybody run any kind of like HPC type workloads with its machine learning or AI stuff on either like Graviton or ampere stuff, either on AWS or have they got stuff done locally at the university or something like that? There are a couple of Japanese facilities, the university that have switched over to ampere based ARM supercomputers for high performance computing. Both of them I believe run well on them. But it is the catch 22 for ARM is that there isn't good equipment for people to develop on. There's only good equipment for people to run stuff in production on, but all of that is too expensive for most people to use. Even with the, with the max, right, like the problem with the ARM max is because you can't run Linux natively on them and you can't do anything meaningful with that, you know, with native workloads, there's a certain class of development that's just completely accessible to almost everybody. Unless there are companies or people like actively pushing for ARM based equipment to exist at affordable price points that are actually powerful. You're just not going to see our move past where it is now. So, you know, maybe, you know, if companies like IBM, Red Hat, Zouza are all like really wanting to have this ARM thing take off. Maybe they should be talking to the people that make these SBCs and stuff to, you know, maybe tilt the roadmap a little bit more in favor of making, you know, high end arm development more accessible because right now it's, it's just not. I'm excited about ARM. Yeah, I feel like it's a bit similar to PowerPC because all that is right now is there's Talos like that's all I'm aware of from the PowerPC space or PowerPC space. There's no other vendor that does anything consumer level. Right. There's a couple of attempts to do something more affordable, but nobody is released to manufacturing anything other than Talos. And like when your lowest tier product in terms of pricing is too grand for one machine, like, that's out of the realm of possibility for almost everyone, like, you know, I very much personally believe that it has been a huge mistake that for ARM development for professional use cases that everybody's been ignoring the fact that we don't have access to good ARM based equipment on an individual level. And this is something that I know I've harped on a little bit in previous OKD meetings when we talked about OKD on ARM. But like I really do feel that unless this problem is fixed, long term, OpenShift on ARM is going to not be as good of a success as it could be to put it very mildly. Without, you know, obviously without breaking any of my contractual obligations to Red Hat, I can just say that like, Neil, the kind of things you're talking about, I have like seen those conversations happening inside that. I can't say anything really like deeper about it, but I know that there are people who are more senior than me, who are like concerned about these things. So. Yeah, I mean, it's, it's even something as simple for me like with this last Fedora, you know, I'm going to break a little bit out of the OpenShift bubble a little bit just just for a brief moment. You know, this last release cycle for Fedora, for Fedora Linux 34, I launched Fedora KDE on AR64. But it was such a pain in the butt to dev and test that I had to do most of it blind and kind of hope somebody would tell me whether something was wrong because good reliable ARM64 hardware just flat out doesn't exist. And that is a serious problem that, you know, no amount of like promoting on the server end is going to fix. You've got to do something else on the developer end. All right, let's move on. Timothy, then Christian. Yeah, we're not the AR64 development working group. So, yeah, all of that we know we cannot do anything about it. We work on the support of the platform that actually exists and for Fedora Chorus, it will still be as USA and whatever and So sure, lobby or there are lobby vendors or whatever, but yeah, we cannot do anything about it. Okay, Christian. Yeah, yeah, I can essentially second that we can't really we don't control the hardware is we're here making the software. I mean, looking at the development over time, I do think there is a trend towards making that more accessible. You can now get these Nvidia Jetson devices for like $400 which, you know, they have 22 gigs of RAM, maybe you can even run OpenShift on them. I don't think they have the UEFI firmware yet, but, you know, it is going in that direction arm devices are getting more powerful and at the same time the prices are going down. And the proliferation is taking place all bit slowly. But I do think in the future, obviously it's not going to be tomorrow. It's probably going to be next year or in two years or even three years, but by then I think we will have the ability to really run an OpenShift cluster on hardware that doesn't cost much more or even less than $1,000. All right, I do want to move on to the next topic. This was a great exploration, though I think it's really helpful in a variety of ways to give viewers a sense also of what's happening and people in the meeting here as well of what's happening in that space. So moving on, our next item is a look at the issues. What are the issues? So looking at issues that are in the OKD repo, is there anything that stands out here that folks want to bring to our attention? There's actually only been one added just in the past like two weeks and it was one about creating a new cluster 4.7 and looks like there was some conversation about it, John Longo. And this is more stuff with Network Manager. We're still sort of bouncing around Network Manager stuff and resolve. And does anyone have anything to add to that ticket? I don't really have anything to add to it. I'll put the link in the chat for anyone that's not there yet. And yeah, nothing really to add. So, okay, nothing in particular in issues have popped up. That's good. Now moving on to discussion items. Is there anything in the discussion items section that's worthy of our attention? Nothing added in the past couple of weeks actually is just mirroring in the repository and certificates. And I don't know if anyone has anything to add to that one. It's been sort of lurking for a while. Yeah, I think I commented on that. I do think it's a user error that is really tested and supported. Okay, moving on now to updates from the documentation group. Mike, why don't you go ahead and cover the stuff that that's sort of in your court right now, or our court? Yeah, sure. I mean, I can talk about these these other things on the on the list. Yeah, I can just repeat what happened in the meeting. Um, linking to the charter from the main website page, I think we discussed that and I thought Diane was going to look into doing that. But, you know, well, we wanted to get the whole group's agreement on that was the just so bringing it to the whole group. Does anyone have any opposition to putting a link to the charter on the front page of the okd.io website? Is there any downside to putting a link to the charter? I think we should probably we should probably I was going to say I think we should probably like do this also if you're like a lazy consensus on the working group as well just if we're really going to try and get like a community consensus about like putting this out there, you know. Yeah, let me get the link to the charter actually because it. Let's see. Yeah, I thought we were going to do it. Yeah, I wanted to run it by the group just use why we hadn't. I'm trying to find the link and it's not popping up. Charter update. Okay, I don't see the link to it actually. I'll find the link and put it in there. But yes, there is a charter Christian can provide a little more context on that. But, and maybe Neil, I think you were there at the beginning. Basically, it's like a page of explaining what the working group is about and people voicing interest and whatnot. So it doesn't sound like anyone. Okay, Neil posted the link to it. So it doesn't sound like anyone has any downside so we can say to the documentation group go forth and Diane. Thereby go forth. All right, straw poll vote raise your hands or say yay. Yay. All right. All right. Just wanted to be sure one thing. I maybe would like to do is update the date that's at the top of it because it has actually been updated. Like, it was updated three weeks ago, but the date wasn't changed maybe make that a p 1.2 and update the date before we link it so everybody sees this is still current and not some some document from 2019. Okay, great. There's nothing else there to do. Is there anything else that stands out with folks that that we should do to the document before making it. Visible. All right. Sorry to interrupt you there. Go ahead and take it away. Yeah, no, yeah, no worries. So the inclusive language update, you know, like the, the O. K. D. the docs dot O. K. D. that is all generated from the O. C. P. upstream. So, like. We had we had pretty positive feedback from the red hat docs lead who's kind of doing that work that they've run the inclusive language stuff on that, but then we also want to run it again. Like some of the linting tools we have on the produce documentation and also on O. K. D. I. O. So like O. K. D. I. O. itself has not had any sort of treatment. It didn't look like it was in that bad of shape, but this is kind of something that, you know, I think is probably going to be coming. So, but that's going on in the background. You know, I think it's being handled by the docs team pretty much. I didn't have a link to Brian put in a ticket with a bunch of notes. Do we want to cover that now or as a separate business item? I put it underneath. Yeah, Diane asked me just to mention it on on the call. She picked me. Yeah, let's do it. Let's go ahead and do it now since we brought it up and I'll just move the link. Okay, so I actually did run the tool on O. K. D. I. O. And I summarise the results in this issue. And I think the primary issue is the primary branch. Obviously, all new GitHub repos now get main as a primary branch existing one still have the master. And so that is one of the problem words that gets flagged by the tool. So that has quite a lot of knock on to the repo. There's a lot of static links where the document links back to itself using the sort of direct links with the branch name in it. There are certain links like the contribution page where there's instructions on how to do things and that mentions the branch. And then I don't know how what the build pipeline is for the site, but I think there's some external tooling that would have to be updated and then anybody with the clone. I think GitHub does a pretty good job of if you rename it'll actually let sort of clones work for a while. But again, for completeness you everyone who has a clone should run the commands in the issue. And just to update that they get config and couple of other things that come out of it. There is a couple of other words like whitelist that's in there that we obviously need to change there are just in comments. But again, we should go in and change those. Other than that, it's pretty clean. So I did put an issue in because obviously there's quite a lot of coordination going to have to happen to actually move that over. And so I'm not sure how that's going to do. But I guess, are there any other repos that the the community owns it on sort of managed by a product team. I'm actually thinking of probably the main OCD. I know there is some some content in that one that probably also needs to be done. The tool is very, very simple to run. And you can run it natively or I actually just use the Docker. There's a Docker file in there. You just run the Docker file and it pretty much works out the box. But yeah, that's we just need to then coordinate get get these issues. Action. Excellent. We also own the the community repository in the open shift arc, which I don't think any automation touches. Which is also the one that contains the charter. So we might want to rename the master branch to main there too. I don't think we can do that in all the open shift component repositories yet because of the extensive automation and CI that that requires that name where where the master name is hard coded in various places. But in that in those two in the community repository in the OCD repository, I think possible. And my sense is the next. Oh, good. I was just going to say Christian my understanding on the individual components was that like those were being left up to the teams to migrate at some point. Yeah, I think what we'll end up doing at the next docs meeting next Tuesday is basically developing a plan of getting all of the changes in in the repos. That we know and have control over as PR's send out a message to the community in our various social media channels, letting them know about this change. And then approving the PR's after maybe a week or something like that, and sort of the our intent has seeped out into the community. And that's probably what the docs team is going to do. That's what I think we'll talk more about it at the next docs meeting. Is there anyone here who's not attending the docs meeting that has any input or anything they'd like to add to this that we can bring back to the docs group? Or is there anyone who would like to join the docs group that hasn't been participating? All right, that's a clear mandate then for the docs group to do what they need to do. Go do the thing. Go do the thing. Go do the thing. Yeah. Elmiko anything else you wanted to say from docs? Yeah, the only last thing was like, I sent out an email earlier in the week about this guides PR that I have up. I think, you know, the last week at the docs meeting, you know, Diane was like, let's merge this now. Like, so I think we're probably going to merge it next week. If people have comments, please go check out that review. I got a few changes to make, but, you know, definitely comments. Welcome. I can put it in the notes for today too. Excellent. Thank you. And thank you for all that hard work that you've put in, which involved understanding a very arcane system actually. It more just offended my sensibilities than anything. From what I know of it, I totally understand that happening. Okay, let's move on now to new business. And the new business that we have is to promote the recording of the open office hours. A lot of people attended it and forwarded it on social media, the link to it when it happened or before it happened. Now we need to promote the recording of it. So there's a link in the meeting notes to that YouTube recording. So if you could share that widely in your networks, that would be awesome. And the next step in this operation is deciding on a date for the next one. So Diane isn't here, but basically she's offered up the Friday or I think that they also have a Wednesday when they do the common stuff for us to commandeer those. And for context, the last one was the second Friday of the month. So it was July 9th at noon. And my thought would be that we do the second Friday in August, which would be Friday the 13th at noon. Does anyone have any thoughts about that? Does that seem logical to try to do it once a month on the second Friday and just follow that cadence. A lot of open to open shift TV does that they have regular slots and I think people are used to that. I think it makes sense to actually say that you could be team will be on second Friday of the month. No, no objections. Okay, no one seems to have any objections to that. If you do, I can't say whether whether Vadim and Diane are able to make it on that date. I won't be, unfortunately, I'll be on vacation, but yeah, if they're there, I think that should be fine. Whenever you say you're going to be on vacation, you have to tell us where you're going to be on vacation. Where are you going to be? I'm going to Italy. Awesome. Yeah. Very cool. Okay, so even if Diane can't be there. I can be there and can anyone else be there anyone else from the group. So that's not just me and, and whomever we try to rope in whomever else we try to rope into it. So August 13 noon, noon Eastern, I should say, so that's what is it 5pm. UCT probably on vacation too. So not as exciting as Italy though. Where were you be going? I'm going to Vegas to see George straight and concert. Hey, it's a Texan thing. There you go. Yeah, just fair, fair warning, Jamie. I was just looking at the calendar Friday that that August the 13th is is actually, I think a holiday for most redhead employees. Oh, is it? Oh, it's like, yeah, it's recharge day. We have these internal holidays and that's, you know, that just happens to be it for the quarter. Okay. Friday the 13th. Okay. Yeah, right. Who wants to give a demo on Friday the 13th of something. Let's see. Well, let's, I guess we'll. Oh, go ahead. No, no, I didn't want to interrupt you. Go ahead. I was going to say, do we want to punt and just keep that date and hope that someone can join me. Or should we change it Wednesday thing as well. There was a slot. Yeah, there is it's the 2 days that open shift commons has slots are, I think it's Wednesday at noon as well as anyone else know. I can check the calendar, but keep in mind that on that Friday, Chris and Bobby will be on. Recharge day as well, so they won't be able to produce, but let me check our calendar. Because that's actually my team. Okay, that was going to be my next question. Right. If no one's there to produce, right? Yeah, he's Chris going to be there. Yeah, exactly. Yeah. Do we want to schedule another week then or do we want to just skip and wait until September? I didn't climb to say that it'd be easier to skip and wait until September. Yeah. What would we want to talk about? Well, that was my next question. It's like, what, what amazingly cool stuff is happening with OKD that we can demo or bring someone in to talk about. So maybe it makes more sense. Sense to be very fun to demo, like even in AWS or something, but that's probably better served waiting until September for a tiny bit of extra progress. If we want to, if we want to do that on Fedora Coro as instead of rail, then September might be better because I don't think we can. Yeah, we're not ready right now. Yeah, so unless you want to, unless you want to demo open shift, which would be lame. No, let's just hack it all together. I think a topic that a number of people have sort of expressed interest in is 4.8 single no cluster. I think getting someone to demo that, especially for a lot of home users, that will be their go to. So I think that would be a really cool if somebody could actually know that. Now, who would we rope in to do that? Somebody that's got it working. I don't know those profiles have made it to. Okay, I don't know that anyone's even tried those profiles with okay. I thought about trying to rebuild. Okay. The scripts they have in that repository. And then I, I didn't. Also, so. Well, yeah, but as I know is actually another profile that's being like put out. I think 4.9 is, I think that's when it's going to be coming. I don't know, but single like a single node open shift is something that's coming as like an edge case. It's a little bit different than I think what we're talking about, but that I don't think that profile has been tested at all with okay. I'm just not sure. Yes, I know stuff is kind of like a stripped down from the operator standpoint like the most minimal needed my understanding of it. Right. I mean, yeah, this is right now it's being specifically targeted at edge. Something I thought about in the past is. Okay, even though it's more of like a sibling thing with open shift. Obviously, you know, can you stuff kind of puts. You know, sometimes like a version ahead of like what's going on even like for in both like for it's kind of come out around the same time but like for nine is like available in the CI stuff even though there are nightlies from OCP. Would the working group like want to try and show but hey we're going to demo like for nine features people so they can see what's coming in open shift but also like you can use this now kind of thing, like for future releases at all or they not want to we want to back off that and let Red Hat have their their promo for those kinds of things. I was I was thinking something similar because there's a feature that I'm working on for OCP right now that hasn't been tested with OCD yet but like it. I was going to recommend I was going to say like maybe we should this could be a possible demo idea as well so like, I think it's total like all these many of the features we're working on are being developed in open source so I don't see there's any reason not to like do that I mean they'll they'll they'll get release announcements when you know OCP comes out but like my understanding at least from a Red Hat perspective is that OKD is supposed to be kind of running out in front a little bit you know so like I think it'd be perfectly okay to you know to talk about these things like big for example what I was going to propose that might be an interesting demo is you know we're getting ready to do this transition from the entry cloud providers to the out of tree cloud providers and that's going to open up a lot more kind of opportunity. We have it working on OCP right now but we could probably try it out with OCD and then demo like how people could do a migration from entry to out of tree and also how they could install a cluster using out of tree cloud providers. So that you know that's a feature we're working on that we're going to you know it's going to be a big thing for OCP but it could also be a big thing for OCD as well and we could certainly demo it and you know talk about it if people are interested. Something like I've thought about in the past something like in the cloud space in the container space specifically I've always wanted to invest because I never got the chance to use them is build packs I know for eight I think they GA that feature. But I don't think I've ever seen like any like written tutorial or video on using build packs within the OpenShift space because it's always like S2I use S2I for everything but now you have this feature from CNCF that is GA in the product but I don't see anyone talking about it but I feel like that would be something cool to that's just a personal thing. But I had something in my head. That may be something good to come from the community though because like a lot of the a lot of the internal talk I see is around using like S2I and using the ODO tool and whatnot to build these like you know source to image. Kind of things like I haven't seen like a ton of people getting crazy about build pack but like yeah like that could be a great community integration to bring up right. I guess we do. Yeah it sounds like we have some ideas do we want to. I mean I guess if we decided in two weeks. At the next meeting that would still be enough time to promote and and whatever particularly if we're going to follow our regular cadence we want to punt this topic until next meeting in two weeks. Okay let's do that and then we'll make a final decision about what we're going to do is if we can get someone lined up. I think it would be good to have a regular cadence and not skip just because it builds that momentum and people know that you're always going to be there. Same bat time same bat channel but if we can't do it. You know then then we won't do it and it sounds like if our technical support isn't there for the producers then we're kind of dead in the water for for next month on that particular day but we'll see. It might actually be worth setting up. I don't know where we put it but just I mean we captured quite a few ideas of what what what we can do just actually let's get a list together of what people think we should cover on those days. And then we can actually start asking for volunteers on who can look at this topic for maybe next month or the month after and start sort of building out a schedule of topics that you want to cover. Well I've got the in the meeting notes. You know there's an item for why don't folks just put the name or the idea that you have. And then what looks like some people have already started actually adding it. So if you have a suggestion for who might be the person to the demo those ideas. If you put it as a sub item of your entry there on just who we might contact for that. Then that keeps everything nice and in organized so. All right, we have five more minutes left we've covered everything in our agenda. Are there any last minute topics or questions or concerns that you want to bring to the table before we call it. I feel bad asking for this. I'm asking just because I've been out of the loop for months now a lot of work stuff happening on my end. Last time I was in the working group there was discussions around operator hub and like the right out registry versus the community registry and the cross poll nation there. I don't really know what that status is routine okay you know cp right now is, has that changed at all or improved since, I don't know feel like this is like November at least Christian go ahead. Yeah, I don't think there's been a lot of development on on that front. We do still plan on having separate okay the catalogs. And I think Vadim has talked to some people internally to to make that happen. And, yeah, but I haven't heard anything for at least three months, three weeks now so probably not since the last meeting that there is an update on on it. Yeah, I will ask him about that again, and hopefully by by the next meeting will have some more new. Definitely it's in the making but I don't think it has a high priority at the moment internally we should probably push on that a little more from from our side. Yeah, I think that's a good selling point is what we have you know in terms of operators. Now, I will say that the document that Christian created the issue was reorganized a little bit with a little bit more context as to what like the three different categories basically of of hubs and so you might revisit that Christian might have the link to it if not I can post it but basically there's a little more context in the in the issue that Christian created that lays out all of the operators that folks are interested in seeing and where they're coming from what ones are internal to red hat. What ones need basically a community developer what ones you know etc so we can start with that my sense is that probably in August will take a more targeted approach. And we had talked in the last working group meeting about starting to reach out to folks at particular companies in terms of having okay D tested in their environments on their platforms. I'm thinking we're going to probably do the same thing with operators is just like start reaching out to folks and saying hey, we would like this. You know, we would like to get some resources we'd like to contribute some resources to getting this operator working on okay D and in the proper place to take a very targeted approach to it. That's what I'd like to do anyway. Alright, anything else. Alright. Thank you very much folks. And again the docs meeting is next Tuesday same that time same bat channel fedora core West meeting is tomorrow at 1230 Eastern. And just will put links to that for the core West meeting folks are interested in attending. And I think we're good to go. Thanks everyone. Thanks to me. Thank you. Thanks everybody. Thank y'all.