 KD Working Group Documentation Subgroup meeting for April 5th, 2022. And it looks like we have a new face here with us today. Larry, do you want to introduce yourself and say hi to the other members? I've been here before. It's just not maybe not in this meeting. I've been around working with the project mostly for the 311 stuff and following along for everybody else as a. Forged ahead into the 4 stuff and trying to figure out my path into that. Excellent. Excellent. And we had Michael V just show up as well. So welcome, Michael. Hello. Sorry, I'm late. No, no problem at all. Do you want to get started Brian with community repo stuff? Yeah, have you posted the hack MD? I think most people have found it. So I've put a link in the in the agenda. We're looking to remove the community repo and so the links there. Mainly it contained five or six documents and that was all. So we wanted to roll that into the OK, DIO website. So I've actually moved most of the stuff and we've got the charter into the. Look at the site now and I put the link where it is. So it lives under the working group section because it really looks at how the working groups work. And however, there are 3 documents left and I'm not quite sure what we want to do with them. 1 is the owners and that really is just the main. It looks like it's just for the main. Working group where at least 2 are the owners and then who can approve. Which I'm guessing is into the OK D repo. And then the other 1 is a roadmap and I don't this probably needs to be taken to the main meeting. But I just thought it was worth discussing here so we can actually take a proposal to the main meeting. It's you get more done that way rather than trying to have a discussion with the bigger group. So my thought is for each working group do we want to. Actually, I think we need to actually have the sort of officers of the working group. So who is the chair who are the primary drivers of the primary positions within the work group. We seem to be quite sort of dynamic in terms of membership. So I don't know whether we want to actually have a members list for each working group or really just. My my suggestion is let's just leave it to the those that have a position. Or some official duty within the working group. And then the membership can be a bit dynamic if we grow. And it becomes where we need more steering committee type membership. Then I think we can move it there. But for now I just say the chair if there's anyone that actually has any specific authority. Like who can who can do a release or who can make a commit to the OK D repo. They should be listed. But the membership is not something we track actively. And road map. I don't know what we want to do about road map. And I think we're largely in the hands of the Red Hatters. In terms of where the OCP product is going. So whether we actually have a road map. Or whether we want to do a road map of like community activities. Where we want to do things like the operator catalog. We want to do projects like the technical documentation. We want to do projects like the testing. Whether we actually have a road map for community activities we do. And whether that's something that we should document and have an aspiration of where we're going in the next sort of six 12 months. So I'll open that one up for discussion. That's a lot. That's a mouthful right there. So let's let's start out with with the first document in that. So well the very first of that is the charter. Which has been updated. Like yeah the last change was in June of last year for the charter. Yes so that has moved. I can't actually remove anything from this repo. So yeah I say the official version of that is now what's in the okd.io. Which is a direct copy of what's currently in there. Yeah exactly. But that may be something that we want to revisit because I don't. I feel like there's a couple things that did eventually need to get changed with that. And ultimately we will want to mention in the in the main starter. Like we have a thing for subgroups but we don't actually like list subgroup chairs or anything like that. Because it says each sub working group must have a chair working as an active sponsor. So we'll have to take care of that in the coming weeks to actually align with the charter. Yeah and I mean I read that I didn't really know whether that meant a red hat sponsor or a partner sponsor or just a community sponsor. Yeah the sponsor part I didn't know what that meant. So I think we need to get clarification from Diane and ultimately yeah go ahead. I was going to say within a corporation when you have a sponsor it's usually an executive sponsor. So within IBM when we say you need a sponsor for a community it does have to be an IBM executive sponsor. I don't know whether Red Hat has the same thing. So that's where I read it we needed Red Hat executive sponsor for each working group. So the diversity and inclusion groups have executive sponsors. Okay different. All right well let's find out and then if we have to get one we'd have to go through the process of deciding who that would be. So for the members here's one thing that Diane and I were talking about though when we were sort of chatting about the membership the one hitch with just letting it be sort of whoever shows up is that if there is a vote on something significant and we've generally there's this hasn't happened because you know we're just growing as a community right now. If there were a vote on something that's significant and someone wanted to sort of stack the deck by having a bunch of people show up and vote that could be problematic. That's the only I think downside to anyone being able to vote and membership just sort of being you know. I guess playing devil's advocate what is the barrier to entry so if you had to be on the membership list the week before I just had everybody that I want to vote into the membership list. The other thing that we get into that is if you have that you then start getting what you need a quorum for a vote and what percentage of the active membership has to vote and then you get into all sorts of areas that get overly complex. I don't think our community's at that stage yet and I would say at the end of the day you and Diane have the casting vote so if you think things are getting hijacked you are the ultimate authority as far as this community is concerned. Yeah and that is kind of what we put for you know for our statement you know for the inclusion statement on the website. So I guess okay so membership I guess then just omitted moving over I mean it lists the chairs but then it's got a bunch of people that you know like Joe Doss I know Joe Doss but he's never shown up to any of these meetings. So I mean I actually think we remove the members list on the owners list now whether we actually put working groups and the officers within the charter or within the description of the working group I don't mind because we've got a section on the website for each working group whether the the officers of that working group get listed there or whether we actually put it into the charter I don't mind but the charter should be should reflect whatever we decide to do. Yeah for sure right and then can we have the read me of course which there's nothing really significant oh it does have the open shift dev that needs to get changed yeah yeah I just said that's just I don't think there's anything there that's going to get deleted because this whole repository is going to vanish that that's the idea that everything that's useful in there should get moved somewhere else like into okd.io exactly yeah I'm just looking at these to see like okay does this need to be now roadmap we don't have a say on sort of the foundational elements but we do have a say on some of the testing or some of the so for example one of the things I'm going to be talking about later in the meeting that Vadim and I talked about when we met last week is I said what do you what do you what do we really need right he's like bare metal testing and and testing in vSphere and have it be automated like weekly to match with you know what's the releases basically and that to me is a roadmap-y type thing to say okay by the end of quarter two or whatever we would like to have a functioning automated vSphere testing either through donations or someone offering their hardware or whatever or at least providing reports they could send out to the group or something it seems to me that that is kind of roadmap-y so can we have something in the roadmap that says red hats you know at the base level there's this will follow red hats roadmap for whatever and then the additional things because like Vadim also brought up cgroups v2 was something that got tested first in okd before it ended up in a place like else so our roadmap could actually include a fair amount of stuff that's that's unique to okd I mean we we could almost do something like red hat does with their their releases so sort of after every major release they do and actually think it's on the 14th of this month they do a what's next right and we could we could do a after every major release of okd we could do a what's next to update the roadmap I like it so I I think sorry yeah go ahead Joe okay one of the things you have to remember is that a lot of the stuff that we're testing first is more because of Fedora core OS rather than anything core with OpenShift itself so we're we're sort of giddy pigs for that or when red hat decides to go to you know RH you know red hat 9 or red hat 10 whatever they do you know pull in whatever version of Fedora they're going to use so you know we're the we're the bleeding edge for that to our detriment sometimes but I mean really we don't do anything else and you know of the major parts of OpenShift you know it's the the sitting on F COS you know that we're that we're the testers for yeah which is is fine and and that may change though with the community operators catalog or with I sorry with the okd operators catalog we may be testing operators that aren't you know available in red hats catalog yet you know um we're in the community catalog it could be I mean I'm just throwing things out that like yeah they may not be directly OpenShift related or they might be a bit tangential or something but I think there's an opportunity for us to have a roadmap but I also think I mean the conversation that we had John was it three weeks ago when we were looking at how do we bring in new features to and and we have the conversation should it be at a major release update or should we do it intra-release I think that sort of thing could go in the roadmap so we can decide when we introduce a new feature like like cgroups version two and it could be put on a roadmap so it's we've got six months visibility of or four months visibility that we're working on that and it's going to come in the next release so it's not going to impact it's not a surprise it's like a yeah we did this update oh you put it on didn't work oh yeah by the way we did this whole bunch of stuff in it and maybe that's what hit you so maybe yeah I think again I think we should we need to take this to the main group but I think the proposal is so actually let me write this in here um so the proposal is that we are going to I've lost it hang on um so I'm just updating if you're in hack and D so the proposal is that we're going to drop membership each working group lists officers I'm calling them officers whether we want to call a different name I don't mind um and we'll do a major uh a next release roadmap so we'll create the next release roadmap so we've just done 410 so we should do a meeting and that will be a special working group so we'll actually have a working group meeting to actually determine the roadmap for the next one at around the time of each major release so that's all going to take to the next working group meeting next week unless anybody has objections or other other suggestions that's great that's great okay okay so just moving things on I'm aware that we're you're muted John was saying something that he was okay muted you're muted John yeah I said the working group meetings have been interesting with the with the discussions last couple of weeks so kind of the core of of what what are we going to be when we grow up yes so I don't know Brian did you incorporate I didn't look but did you incorporate anything off of the google doc into your stuff yet um or do you want me to or how did you want to move forward with that the stuff from Vadim yes yes I've been trying to pull everything together I'll get on to that okay so folks yeah so for folks that for for a little bit of context in this I met with Vadim Friday for about an hour we had a conversation about sort of as I said future needs and the future of vocation in general Vadim anticipates being out for at least a little while longer and so you know basically if we find stuff we can put in a merge request and either him or Christian or there's a couple other people that are listed there that can approve those um merges in terms of where what we're going to be when we grow up and I'm just paraphrasing for Vadim because he's not going to be coming to any meetings in the new um his thought is that it would be too much to take on to do all of the infrastructure and sort of go outward from Red Hat because there's a lot of like interaction that happens between him and Christian and the other Red Hat teams in terms of features issues and whatever and then we would sort of be on the outside of that if we sort of split off completely into putting in you know our own main repo and stuff like that so his suggestion is that we sort of fill out these edges of testing and stuff and get infrastructure in place for vSphere bare metal testing and things like that and just keep sort of walking along with Red Hat and improve communication and try to get the ear of more engineers make the case that we need the ear of engineers more often to get things tweaked that we need so that was that was his perspective because I know at the last few meetings there's been discussion about well okay should we just have okay deep sort of split off and and sort of be its own thing and and Vadim's perspective was that we shouldn't do that and I was actually kind of convinced after talking to him that I would agree with his perspective but I'm curious what other folks would think about that case of the argument you made I agree with him only because we there's so much that happens behind the scenes and we don't have we don't have the manpower infrastructure to take that on ourselves you know we're not you know we're not the Kubernetes group you know and stuff like that you know honestly there's what six or seven people you know that are actively engaged we don't we don't have the resources what and what I would like to see though is some of us get a little bit more access maybe I mean you know take a little bit of the load off of Vadim and Christian you know you know I was a you know people who are active enough you know maybe we should have access to be able to do things like you know maybe not do the commits themselves but have a certain level of access to be able to do some of our own testing and experimentation you know without having to bother Vadim or Christian and stuff like that you know be able to have almost developer access but not administrative access you know to those repos so that we can do the testing and stuff because a lot of what we did this past weekend I was dependent on Vadim to do because I couldn't do it myself right yeah but I don't see it I don't see us getting away from Red Hat but I mean my take is I agree and I think it's interesting that some of the comments have been made about us becoming an officially upstream I think that that opens up some interesting possibilities but even if that doesn't happen one of the challenges I have is as a community we don't actually know what's going on inside Red Hat now I know a lot of OKD is linked to OCP strategy and as a as a commercial company Red Hat have a strategy of how they're going to drive that project forward driven by customer needs and commercial needs so I get that but I mean and I'll come on to this I've been trying to write the technical documentation and what I'm trying to do is do a guide of let's make a simple change and let's actually go through the process so rather than just writing abstract I want to actually have a worked example within the documentation and I found even looking at a simple project I tried to do a very very simple change just change the documentation link in the console it's not a custom resource it's actually built into the console operator and I thought that's a simple change it's very visual you know immediately if you've got your version or the official version because one's going to take you to docs.okd.io the one I'm going to change just goes to okd.io so you see immediately when you click the documentation link which operator have I got running and we can't build the console operator the actual build image is within an internal Red Hat repo that needs authorization so I actually can't run the docker build or the podman build command and so the build image is behind that registry the actual base image is behind that registry and the make file uses the build image so you find that even if I wanted to and then you say well we could actually do a pull request as a developer if I can't actually build it and use the same tooling and the same base image that the the the CI system will use I'm not comfortable doing a pull request because everything about me tells me that you should never check in something that you haven't tested and if you can't actually if you don't know what the tooling is you don't know what's in the base image that it's built that it's bundled with I don't feel that I have enough access to be able to do a pull request so the funny thing there is is that that was actually supposed to have been changed because there are community versions of those red hat build images but they've never done it ran the same issue when I was doing mco testing I updated the image the docker containers myself to put in the proper image and then I was able to you know to continue forward but yeah it's been mentioned multiple times and they have not made any changes to it so yeah it's like if you want to do this here's step one you have to change your build image so you use the you know use community available docker container image and and that's I mean so I think that's one thing that we can document what is that community image because I couldn't find it I started digging around key.io in the open shift and there isn't an image with the same name in that in that registry and there's was like 69 pages of images in that organization so it's like I just have no visibility in terms of where I go so I think that there's issues like that I mean and we don't have access to the prow system I don't even know whether we could actually recreate that if I have enough resource because again I can't see where the config for that is so if I wanted to set up a mini prow obviously I'm not going to be testing against all the architections and platforms that the Red Hat system did but if I wanted to create a cut-down version that took around the basic CI steps that's in the release repo I don't know I don't have enough information even to do that and I also found out that several of the repos have out-of-date information in the read me so if you look at the read me to say how to build it you will fail to compile because information is out of date now yet we can start putting change pull requests you know raising issues to say well this says use golang 15 but the actual repo uses features that can't be used before golang 17 or 16 or whatever yeah and there's actually two build pipelines that that are in the process so I want to we're sort of at halfway through the meeting yeah I want to make sure that we that we just determine a plan forward for this so let's let's um let's see where we are here in the agenda let's do let's go back to the removing the community repo part so it sounds like we have a plan with what to do with the stuff that's there we will rework the roadmap to incorporate stuff that we are going to do specific to okd and that will be an interesting discussion because we'll want to revisit that roadmap on a regular basis with this team and with the the general team you know quarterly or whatever right um so let's move on now to website styling so I put the link in and so Brandon he's a contractor working with Diana within red hat he's taking a look and he's made some changes um mostly the changes are on the lights not the dark and it's it's significantly easier to read so I'll put the site there and he's got a couple of things he wants to check out and then he'll be doing a pull request so again if you look at hack md the link is there go have a look have a play if you see any issues let me know and I'll bring them up with with Brandon and we can get them fixed but that is what we're likely to push into the new and onto the website as the styling similar colors similar theme but you notice that the background's got lighter so he's made the the main page lighter and the side menu is a little bit darker actually he's flipped out he's flipped those those colors round and things like that and so that's really all there is there there's been progress made that's the the sort of proposal and that's what will be pushed live unless anybody feels there's an issue it's definitely more uh legible yeah yeah it looks good it looks good all right does anyone have any qualms about uh brian moving forward uh with those uh and approving the pull request uh that's going to be made hey brian I put a link in the chat with uh issue that we have with the build images okay great thank you can reference that so uh so by the next meeting you think by the next docs meeting brian that'll be the merge of i spoke to him yesterday um he said he had a couple of things to check out and then you'll do a pull request so i'm hoping that's going to be within the next week or certainly within the next fortnight i'd hope it'll be live okay uh modified links and then brings then brings us up to the next one at the minute okay the IO changes are blocked um tinnithy did a change um which was something to take out the license which was a duplicate so that was good but he also did a change to on the main page to the mailing list he changed the google groups direct link to a red hat list dot openshift dot red hat dot com i'm not familiar with that site but it's giving a 40 it's not a bit it's not there yeah so that is yeah there's a um that's where they did that there is an open shift mailing list um and it is hold on one second i can actually probably find it because the first of the month you always get the um uh let's see the sort of notification on it there is a mailing list i've never seen anyone post on it uh oh here we go mailing mailing right uh open shift mm is one of them yep that's the one he said yeah let's hold on one second let me see if i can get the link to it here i've i got my email in front of it let me go i mean the question is do we want to go to the open shift because he's actually replaced the google group which is our our okd mailing list for working group members he's actually replaced that with the open shift mm yeah and i don't i don't know what what the motivation for that was because actually the link even for for the settings for my notification email that i get received on april first doesn't work either so that site i don't even know that the mailing list is active itself so yeah yeah i i don't know whether it's just a red hot site that's down but it's been there for 22 days and obviously it failed when it got posted because the link is still on the site is still the google group yeah so um i don't know whether it's ever worked but i was just wondering do we it's at the right change to actually replace our google group with the open shift mailing list no it's not i don't know we should put it back to google group but i can uh intimacy made that change yep oh that's weird okay um so i have um stuff that comes in on that and they um i have that on an archive the actual list that goes in there is listed um lists open shift at redhat.com um that archive actually i have an email that says turning off legacy mailing lists so it's interesting that uh it points to the same location as the old okd 311 um that they said they're turning off ah yeah so so it looks like that that site has been turned off yeah but i'm still getting mail from it right same i'm still getting at least the notifications yeah the notifications yeah larry are you getting any actual content i just get the the monthly notification from it i the last one i got was from august it was an archive saying they're turning it off okay and that was from august of last year okay so that seems defunct so no we can revert that change okay i'll put that back yeah right i'll put that back to the google group i mean if someone should mention i'll mention something to him at the fedora quarter west meeting tomorrow um that hey just so you know this is a defunct mailing list yeah okay that's that's easy fixed i mean the next one goes back to what we were just talking about um i put a link i started writing up so um i'll nico put a link to a couple of repos that had to read me and some documentation on how to build modules and so i've been pulling that together with some of the stuff from um the the note that jamie and vadeem did um so this is where the the technical document is going to live i'm actually doing it in my fork of the okd.io just so we can get people to review it and feedback on it and put ideas in what i'm trying to do is i'm trying to do this worked example so we we we actually talk about what or we can actually show how to do it because i think with a lot of this stuff the first time you go through it it gets quite confusing in terms of all the pieces and how to actually go in make a change and as i say i try to pick the simplest example i could think of using the console operator it's very visual you know immediately if it's worked or not um but then you can see um and if you look at the documentation and actually back i can probably bring up one screen if you look at the documentation it it actually is um it is actually got a lot of um stuff in there um so showing that screen hopefully you can see that yeah so yeah if you look um there's a lot of questions um we've actually covered this like we can't get access to that registry um it sounds like there are meant to be public images available um so we actually need to work out how how we can do that um just doing some to-dos um a lot of the docker files so again if anybody wants to comment on this um either comment on on my repo and raise an issue on it or we could actually start a discussion group but this is where i actually going to want to start writing everything up um a lot of information that you that you got Jamie um is very useful in terms of documenting the release process again it sounds like you need access to internal systems to actually be able to do a release but at least understanding how it works um where the repos are i'm also going to have a section on the operator hub so as well as the the core functions in terms of the add-on functions like operator hub um um and again i actually haven't found the source of say like the registry the red hat internal or sort of the red hat um catalog on ocp so i just need to work out how all of that works and where they where all those little bits come from um and i'll document that as well and then hopefully we are going to get the ocp registry um sort of the okd registry built into this one as well so we can document that as that gets built up yeah and apparently the scoop on that is that this bug that was uh there's you know this bug that sort of hung everyone up stopped progress on the okd catalog so i i see there is now a repo there is an okd catalog repo now in the open shape org so i'm guessing that's where it's going to end up um i mean the other thing that i would say with that is um i mentioned it in the last meeting that um the community repo was pretty much bricked on okd because any community operator that has a dependency that red hat have a supported version for the community catalog will list that red hat which is in the repo that obviously okd doesn't have access to so even things like eclipse che the project key they have supported dependencies which are in the red hat registry on ocp so they actually get broken um on okd so you can't install che um it says it's missing dependencies the workspace version two the defile version two operator is is required but that's in the registry we don't use the upstream one so i'm hoping that the okd catalog will fix the okd community which is identical to the red hat open ship community repo um that's really it so if anybody wants to help out comment join in um i plan to do those and and it really then comes back to what is the purpose of the technical docs um if it's if it's for hacking purposes then we can say just substitute go and use the the ubi eight images your base image um and we can say is that if we want to actually want one documentation then we're going to have to find out what is the official image that the build uses and we're going to have to provide guidance on how to do that so you can actually do a credible build of an operator and test it before you do your pull requests i commented back on that uh issue on five fourteen uh just saying that this is still an issue and it's actually a blocker for basically anybody in the community who wants to do um yeah builds without having to hack you know really the docker file um but that is really the the way to get around it at this point um is to hack that now maybe what they could do is have you know docker dot community or something in those then you would build with the docker dot community or something like that but you know if they want to do something internal but again from what i have seen so far there is a lot of um pushback on letting people in the community do hacking on various components within open shift and i think i think this is this is part of the problem because it's not upstream it is the commercial product and then i can see from a commercial company that this is a licensed products and and the repos although they're in the public they are for a licensed product so um i can see there'd be some sensitivity there um in terms of how we do that um but it goes back to the point i think that bad email is that we don't want to break from open shift because the whole point of okd is it's the community vision of open shift so anything i do on okd will automatically work on ocp and and if we break that link then i think we lose a lot of the value of what okd is about mm-hmm yeah i don't disagree and i think the technical docs you could also add as a bullet point there um uh or also for testing to heaven so yeah contribute to okd via pull request but also just test release testing i think you could add that as an item and and that's sort of um you know aligns with the discussion you know about like okay we need bare metal we these here etc just for testing even if you're not like making changes just to test that releases are working well if anybody wants to put it on i i think $15,000 i'll build a base of a bare metal environment in my house but i'm i'm just yeah i mean well here's the thing is you know vadeem was actually talking about he uses a laptop he actually bought one of his old company provided laptops because apparently you can get it for cheap after its lifespan you know uh and does his stuff on that you know i i can donate some some hardware cycles to it i've got some some machines i can donate i think the the thing is making it official and actually getting feedback from folks that can donate cycles of hardware that they have or or cloud resources or whatever it's just documenting that and making sure that the community has access to testing results but i i think the bigger issue and i think where we need more technical input is automating it yeah not making it a manual process having it repeatable so someone could actually run it either on a schedule or a command and to know that if it fails it's not they didn't mistype something or have to double check themselves but also if they don't necessarily have the skills they have the determination the will to want to help but they don't actually necessarily understand it all having something that is automated can often be a good way in and then you start digging into the automation to actually find out what's going on under the covers so i i think having now i'm assuming that red have have all of this in their proud configuration because they do it for ocp so again it would be nice if a red hatter could sort of unless there's anything proprietary in there that they couldn't release but give us some starting places of this is how ocp does it maybe on v-sphere that would give us a starting place now maybe we have to migrate the script the prow runs to run on i don't know tecton or jenkins or whatever um because that might be easier to set up than a full prow so brian before oh so let me brian um before you got involved shortly before you got involved i actually wrote a suite of scripts for doing v-sphere upi and then recently modified them to do ipi and it's completely automated um there's a repo it's called oct open shift uh i called it open shift uh cluster tools or something like that oct um i can add that and walk people through it to the docs and we can get people automating testing on v-sphere ipi and upi um i think is there a free sorry no go ahead is there a free version of v-sphere now or is it is it still licensed to actually there is a free version of v-sphere that it comes without the basically it's it's you you only get the web control over individual nodes you don't get control over all of the nodes in one interface at least that's when i checked last year that's where they were at um say no cluster yeah right right and would that be sufficient to test okd or do we really need the i think it would be for for upi in particular because upi um you know you're you're you're you're hard coding your ip's if you want or you're doing it through an external ghcp server that's on the network or whatever so you don't necessarily need all of the cluster pieces of v-sphere um to do that you know but we should set up like a hacking oh and this came up with dianne was we need another hacking session basically right like we should we did one i think like a year ago almost like and then so it's time to do another one you know like maybe on a saturday or something like that and have everyone and i think i think this would be a great this would be a great thing either with libvert or v-sphere let's go set up an automated test environment um get it running then get it documented so we have a one-liner command to run go get the latest nightly or release and run the test pointed at a an over to a v-sphere or a bare metal system set of systems i probably need to do a little bit of work to bring it up to date but um i think i use a basis for the scripts i'm using for my automation um okay so we have about 10 minutes left it sounds like we've got a plan for that as well to start formulating some automated testing um i created like a dummy repo in the okd project github just to play around with stuff and and upload um do we we talked about this a little bit before do we want to do the meeting minutes as plain text or do we want to throw them in as formatted pages does anyone have a preference i think we want to put an mk docs over it so just make the markdown and then put an mk docs on top of it i actually think during the meeting though we should still use hack md yes yes the problem with the git is that you can't have concurrent editing so i think we should have hack md for the while the meeting's on and then just because that's markdown anyway we can just copy and paste that into a exactly so you want me to start doing pull requests then to add it as pages to the ok di or how do you want to do this um i mean i think pull requests are good for source code when you want to do reviews and have a process around so you don't just kick it in somewhere like meeting minutes just directly commit it okay all right it's up to you i i mean i this is sort of your baby so i'm i want to be respectful of you've done the most on this so i mean i i'm quite pragmatic i i think for things like source code where it can if someone screws up it can break people i think we want a a more robust process like a pull request i mean for things like documentation especially meeting notes i mean why do we want to go through a pull request and have to have multiple people involved in you copying from a hack md into the the repo to me that's a let's not make it any more complex than it needs to be i'm happy if someone says no let's let's be a bit more robust about things but i'm quite pragmatic i think documentation okay i will always i will always ask first before i make assumptions about just throwing something in just out of respect uh do this so are we can we should we do a a can we can we do a nested meeting minutes link so that it's like okay the working groups okay the working group meeting minutes or do you want it like a link in the working groups oh actually it's all like okay so the the main working group page i guess i could put the meeting minutes in there like a link to it or do we want to do like a sub menu item well it's going to be a separate repo it's not going to be in the okay the main doc main repo so i think there's going to be a link and we'll just put in mk docs wrap it around that meeting one so there'll be a github action that when you do a commit it'll just go and run the thing create a new copy uh let's see what else do we got here um so there are a few excuse me for being probably being silly but is there a link to the github for um library had to modifying okd is there a github repo for that if there is i don't know what it is and i probably yeah if you go to the top of the website there's a github link at the top right for the document to the repo okay yeah because i'm looking at i can see stuff i can add right now once you put your modifying okd stuff back in like how to hell to build a new so so that's on my personal one that's not right that you're going to get to the openshift cs okd.io repo because that's the link in the mk docs and mine is in if you go to github.com slash be in this and it's okd.io so just change the url ramp at github.com can you throw that in the uh in the chat real quick yeah it's in the meeting notes too is it yeah that's the web that's the that's the rendered so it's github.com because i'll fork that in all right some of the stuff because i had to build multiple releases over the last couple of days with the issues i'm pretty i'm pretty confident at it right now that's good uh and the other thing is i think um we should um sort of troll the slack channel for things that look like like i really want to help michael kramer this guy that's really been struggling like so stuff like that i want to look at stuff like that as opportunities to improve documentation right yeah so well part of it i think is is internal to him too but but yeah yeah but there's been other people on there too who have struggled with a couple of things and some of it i can help with some of it i can't anything that's bare metal i don't have the resources to help with the whole thing with uh with all the um uh machine config operators not being up to date that's uh that's when there's biting a lot of people in the bite and i haven't figured out how to recreate that easily um michael did you get a um uh an issue i think it was two days ago for some um uh product documentation there's some stuff lingering that's like references to uh rel uh and whatnot in the okd documentation i think as a matter of fact i think it was submitted by michael kramer was there pr yeah let me see if let's see if i can find it um yeah i i cheat i i i use overt rather than directly but it just makes things a lot easier gives you a nice admin console and happen to that chat message um where is it someone and i said to put one in and they said that they did submit an issue issue was created um yeah he said the issue was created but he didn't link me to it um and basically it was that on this page put it in the chat um this the uh referencing open chest for silky and fedora or or yeah it was like it's it's kind of a mix of of things that need to get changed great um and i don't he didn't give me the link to the pr but i can search for it after the meeting uh and or the um i get it opened yeah see this one here well i got it need for subscriptions and stuff like that and there there were a bunch of those i think there's probably a lot less now but you know being created from the red hat dogs and that stuff happens cool thank you michael all right we have two minutes left is there anything else that we want to cover in the last two minutes that we have the meeting no brian i'll send a poll you're away you're probably in a day or so with some changes brilliant great thank you yeah thanks for your help john thank you brian thank you larry for showing up and providing some background on mailing with stuff that i was scratching my head and i'm sure other people were scratching their heads about as well so jamie i'd probably need to communicate with you directly i tried the OCT E your OCT project and there was some idiosyncrasies that wouldn't allow me to use it okay absolutely yeah it's it was a labor of love and then um i switched jobs and it sort of languished a little bit the past year or so and there's probably some things that i had of assumptions that i need to to code around uh moving forward but yeah okay uh so we'll call it now let me stop recording thanks everyone for attending the meeting and next week is the general group meeting and then in two weeks another docs meeting about to record him