 So welcome to the docs meeting of the OKD working group meeting. So hi Eric and hi everybody and I'm going to put everybody on camera today just so that we there and I'm sort of working through the recap blog post. I've got a preliminary one in the okd.io blog this morning and most of the main stage videos have been uploaded and separated into their individual talks so there's one by Charo, one by Vadim, one by Jamie and one by Joseph. So those are there now plus the one long full on edited one and there goes a dog behind Mike. I love this. I don't have a dog anymore so I'm living vicariously through everyone else in zoomlands dogs these days. So that's happening. I just had and he won't be able to join us. Jamie from Umish has another obligation at this time but I just did a chat with him. He's I have run completely run out of disk space on all of my hard drives and the sessions, the track sessions were very long so I managed to get the homelab one edited enough and then ran out of disk space completely. I'm now so it's been one besides losing internet access I ran out of disk space on my editing machine. It's been a fun couple of days but that said I got the homelab one up. I just retweeted it and there should be a blog post shortly on okd.io. I just have to add in the links to the individual videos as they go up and Jamie from Umish is going to edit the other three sessions and upload them to YouTube for us and I am forever grateful for that and you can see what we're doing. The question that I wanted to discuss today and if other people have things on the agenda let me know was now that we've done the workshop and there's a number of stubs and Eric and Robert I don't think you were there this weekend. I didn't see your names on the weekend workshop and I'm not going to hold it against you but Mike if you can share your screen and show walk through the docs on the deployment and configuration guide where they are now so that Eric and Robert are up to speed on that. What I'd like to talk about is where they should live and now that we've done that and what else we need to do to them to kick off this and then the other thing I'd like to talk about is reviewing all of the docs that we have after we have this first conversation and trying to come up with a cohesive coherent documentation strategy for okd and I've had a lot of coffee so I'm going to try and not talk too much today which is impossible because it's me. So we did have a few pull requests against it I believe during the thing but I think it was all from known entities. Did we have any at all from attendees? Yeah probably not so I'm sorry I just need to unmute my mic there sorry yeah no we didn't it was just Shree and Jamie and Padine put up PRs. All right and I think the one that Charo did if you go to it is a good example Charo I think just put in a stub for his home lab that reached out to his blog post. Yeah I don't think he created a stub and he didn't create a PR yet I saw that oh boy now I don't know where it is it would link to me like a gist I think he had put together but yeah no he didn't he didn't put a PR up for here yet. Yeah so what I'd love to ask you to do Mike is if you can put in a stub for a home labs because there's and there's so that we have a stub there and because we in the workshops we ended up with I think three different approaches to home labs Craig Robinson's Shree's which is very specific to his work and Vadim's so I think that was I just noticed we got two bare metals there I need to clean that up a little bit so did you want to see like a top-level home lab okay because what it seemed like what people had been doing so far what I tried to encourage was people were going into bare metal and putting their you know putting their username with their deployment. Yeah so um yeah good yeah so that's probably not like great documentation process so there still needs to be some cleanup and editing on this on these configuration guides so that they have some uniformity to them and and to be quite honest I'll say what I said to Jamie earlier I we were looking at the OpenShift.com blogs and the home lab for 4.5 that Craig Robinson did that I reposted and OpenShift.com head is still in the top 10 of blog things that get hit on OpenShift so the wording the wording home lab is an attractant to developers and more so I think than bare metal um and I think bare metal is more along what Andrew and Justin Pittman did that it might be there and I think he also has some stuff to give us to track down but so I think this is this is the link sorry yeah yeah so that's that's Charo's stuff so we have some still some work to do here before it's ready to pull into another repo and make it live and breathe um yeah so my my druthers is that the bare metal one maybe get extended to be bare metal with I think he used live vert um yeah so that when you said that yeah like uh Jason or Justin and um you know blanking um yeah they're and Andrew um they're their lab yeah I wouldn't have called it bare metal so much as I might have called it live vert you know because that's really they were exercising the live vert pieces of the of the cluster to build there now did did they use overt uh this time uh as on top of that or I didn't get to see that because uh I was being the straight man for Charo yeah um I did put the raw file up on um YouTube um so I can share that let me just grab that mistake I I thought it was I thought they said live vert but I I mean I also joined late because I was at that session with you know Bruce and Charo yeah so let me just um yeah asking good questions yeah so I'm just gonna I was trying at least I'm just gonna go back in the chat and I'm gonna put a link it's so it's so don't share it quite yet everybody can take a look at this is Andrew's and just skim past the first 10 minutes because that's all them trying to figure out how to set up their screens um and that's what has to be edited out but um that's the video if um Bruce you want to look at it or anybody else wants to look at it so Diane I wanted to step back for a second though and go back to talking about the configuration note your guides uh repo because I agree with you like you know the more I was thinking about it over the weekend and the more we're kind of like you know we started to get some pull requests in there you know yeah I one thing I would love to have is like a template that we could use we could kind of say here you know and and I'll look at the sections we have now and maybe I'll just try and call out those sections and say you should have like a hardware section you should have section there um but then also like really the shape that this information is starting to take is yeah it's much more like a community response it's like here's how people in the community are deploying this but I don't really see it resolving into like a set of like oh here's the one I pick and like I'm going to follow this one exactly because right now it just feels very much like a you know like just community notes that people are throwing up there and saying here's how I did it you know so if there's a way to make that more useful to the community maybe by organizing it better or whatever like I you know I'm all yours for hearing about how we can kind of push in that direction was this two things that I wanted to say about that one is um is is you're absolutely right um and a template like in the readme file at the beginning of the configuration guys what you know what should be in your your stub or in your guide and then to figure out the categories better so bare metal should maybe Andrews and um Justin should get shoved into something called I think it's live for if you watch the video you can see it there um and I was also going to say this and everybody can argue with me about this is that I think it should live in the okd.io site um rather than OpenShift Origin or in okd land and I'll explain what I mean by that is one is if it's in the OpenShift repo it um has connotations of officialness um and it also means that we have to follow their guidelines for merging new content in um and ask their ask the engineering team that owns the repo or bug Vadim or bug Christian whoever has merge access and um frankly I don't even want merge access on the OpenShift repo at this juncture there are you know thousands of customers relying on it and if I screw up or I don't want my name on the front page of the Wall Street Journal for introducing some bug um however on the okd.io site I do own that um and um there are a couple other folks there so if we move this and then we could format it any way we wanted we could make it as beautiful as the porter.sh site um we could do whatever we want so what I you know and and I would not that I don't love you Mike but um I would love to move it over there sooner than later um so that people get into the habit of pushing stuff there if that's okay and yeah I'm totally happy yeah yeah and so um I'm gonna pause for a minute and Bruce you were about to say something so I'll shut up for a minute and then I'm gonna ask Eric and Robert because they're the newbies on this call what they think about it I was gonna say wait sorry not to interrupt Bruce but Robert had a question in chat too so maybe after Bruce let's pick up Robert's question okay well he can go first if you want I'm not you know standing on ceremony okay um so Robert had asked like what's the difference between the the guides repo that we're talking about now and kind of the official installer repo where you know you've linked to the upi so that those things you link to in the installer repo those are actually artifacts that I believe are used during by the installer to actually create the installations what this a really repo that I created originally we were just calling it notes because what we wanted to do was we wanted to collect information from people about what their clusters look like once they were actually deployed right so we wanted people to contribute like I know I'm running vSphere and I followed the standard you know IPI installation and I ended up with this but then maybe that that person says but I also run vMotion in my cluster and I have all my storage set up in vMotion so that I can you know move disks between nodes or something right and then they might have a section in their thing about like here's how I set up vMotion right and so the idea was to start getting all these user stories for the different ways that people are configuring their clusters kind of I don't want to say day two but once they get to day 1.5 you know once they've passed installation what does what does their configuration look like and are there any customizations that they've had to do you know to kind of get there so that was the intention on on creating these guides and just and so those and those are maintained by the red headers and the engineering team and not us for the most part we can make if something's wrong or grammar mistake it's linking or what you can make a poll record you because you're it's all out in the open but the stuff that we own and the stuff that I can control or give Jamie and other people in the community access to merge into it lives here in this github repo and if you're wondering what the CS stands for long time ago customer success so and it's also where project quay lives and the openshift commons website lives and hopefully soon commons will get lifted and shifted out of there and moved into if somebody's merging migrating to Drupal it feels like going back in time what is that movie with where you go back to the future part 21 that's like I'm going to a Drupal site yeah okay but uh hey a long time ago I was a big player in the Drupal community so hey I'll get to meet some old friends again so that's all good but um but this is where I think we should lift and shift the guides to like we have the blog the config community configuration the end deployment guides um are here and that way um and I'm going to keep putting in the porter dot sh site is my um fantasy island um site for really good documentation and I don't think I can wangle enough permissions for everybody in the community to do this under the openshift um re open source repo I I know I know because I own it I can do in the okd.io resource so that's and and that will also especially and this is what the way that Diane thinks is if we put in a lot of stuff about home labs like we have you know one of the guides is about home labs that will help drive traffic to okd.io um and get people more people involved in the working group um because um to be there are there are I would say probably just shy of 500 um people who have deployed openshift um okd for somewhere in the universe whether it's production or or edu use or you know or just your home lab there's there's a good chunk of people here um and that that's awesome um and and there's about I just under 400 people on the okd working groups so the thing is to try and get them to come to the okd.io as the portal into that links to everything and um and I think we can do that um and I think that deployment and configuration this is why I've been so hot on these doing the okd marathon and then do the these other things and getting up on youtube um is to keep driving awareness of okd for and it's been semi effective um and probably as effective as the sales team at openshift wants it to be um but it could be better um and it could be easier to to maintain so um what I was going to and and now I'm going to stop for a minute so Robert and Eric you're guinea pigs here today I don't know where either of you are from um what companies whether you're red hatters or not but um tell me what you think of that idea is that where you would go your I see the red hat sneaking out there Eric yes um behind you yeah I have one over there I'm just tilted the wrong way today um so tell me what you think is that am I thinking correctly about this and you can say no because I'm very good about say hearing you know so um what I like is the idea to summarize all of the okd related content on the okd website I also like that we have this new word okd for the free and open source part of openshift because it is it sounds very different and there's always this kind of confusion between the product and the actual open source project and having everything as okd and getting the okd letters more into the face of people will also teach people more what is actually the difference and what is the upstream project basically and um yeah with the documentation I I feel like there's something where I want to help so I'm personally very interested in the single node areas it's also why I want to look and I was also in the workshop with Charo and also afterwards we had to have a call so he could help me get the last steps done and I also made some changes to his scripts and would have some small changes that I would like to make to his blog post where I don't know where I can hand these changes in and having a central place where we put all these documents is very helpful and yeah it's it's really a little bit a question there's the open shift documentation that's on the open shift website right then there's some things that we are now doing for the okd stuff and there's also if you have a customer account you have some customer documentation as well right and it's also not one on one and there's also a little bit a difference in focus between what is on open shift on the public open shift website and what is basically the customer documentation where you need to have a subscription form and it's it's probably a good idea is that we have a clear understanding what we want to have in each and if I understand Mike sorry Mike is the correct name because it always says Emiko so I'm always not sure how I should go either one works either works then I then I call you Mike and if I understand you correctly what you basically want to show in the okd documentation is like what are what is the result of what people are making out of the documentation right and there are also some individual adaptations and to present different scenarios basically like in the past we also had these kind of deployment scenarios that like a solution architect could bring to a customer that we maybe have something like this as a focus for the okd page that sounds reasonable for me yeah so no big no from my side at the moment okay so Robert what's your perspective are you with red hat as well I don't see a red hat behind you so um I have no thanks so yeah I have a red hat from a conference but the girls are playing with it yeah it sounds very nice but I'm afraid how we can update or again the documentation and up to date state the whole time maybe it's maybe it's a good good idea to put it like a block because if it's like a block you know the date you know how old the stuff is and you know it's one year old it doesn't work maybe yeah so but but if it's put in a kit repo it's not that clear for which version it works and works not so I think Mike in your template that you're discussing much like with the blog posts um that um we have it has a date timestamp on it like it that's one of the things and you know it has the ability to tag so what if you look at the read me and I'll just share share to do something like if you can if you don't mind me asking you to do this is to to create sort of a template for what the should be in the guide um because Robert you're absolutely right these things one thing that I really despise is is documentation by blogging you'll always hear me rant about that because internally at red hat we've had this I I've been at red hat now eight years and we get a lot of brownie points and pats on the back for writing a good blog and just like Craig Robinson's home lab 4.5 blog is in the top 10 it's now completely out of date or not completely but just enough to make it nasty for someone trying to do it so um if we had a revision scheme and a freshener state in the um in the documentation I think that would help in the in if it were in the okd so that people would it would trigger people and instantly to go this is release 4.5 this is the version of it and then you could have sort of a bit of a hierarchy of that and um and then we could um what do you call it when you retire something um an older version yeah so we could retire deprecation thank you very much for that word today word of the day deprecation I want to deprecate myself um for a week and go on vacation um but I think that Robert that's a really good point um and as Bruce put in a link to yet another um set of documentation red hat is famous I think externally for like really complicated strategies for um in every business unit within red hat even though externally everybody thinks of us as duck red hat and thinks we are totally coordinated on every product silo um you you will find stuff like that Bruce he's he's linked out to the red hat dot com sys admin kubernetes cluster laptop um link there and um which is you know another resource but that is um I think it sort of worked yeah and is that one OC I didn't open it up but is that ocp and someone who has uh no no no I think that I think that was no kd1 um but uh it was back in the time of beta yeah um but but I mean if you look at the architecture that that he's using there um it's pretty similar to other guides uh but so maybe his I'm just taking a look at it so as a good template is that what you're suggesting with this uh well actually I guess I had a couple of comments uh although you know Eric and Robert you know made a number of excellent points uh in that uh things do quickly stop working and um you get to where um often something if you tried to follow it exactly it wouldn't work for a variety of reasons so for instance uh you know Charo has some excellent guides um most of them don't work because uh they're they're based on versions of centOS which you can't get anymore yeah okay uh so he's he's uh his cluster version um was based on a uh you know centOS which had a uh centOS minimal which the last centOS 8 version does not have and then that breaks everything uh and so he for about last year he's had a uh a note you know going to centOS stream maybe only six months but you know of course everybody's busy and that would be a big change you know like I'd be happy to have him help him work on getting it together but now he's getting paid by Red Hat to do actually useful stuff so that's going to impact his time presumably yeah but his wife is away for a week um and he's been editing everything so um we take advantage of when our partners leave um and and that's what I think uh just say I think one of the things that uh could I've actually tried a bunch of of those uh and uh sorry the first one that worked end to end was was Craig's 4.4 uh one and I actually liked his architecture better than his switch to 4.5 you know but uh and none of them quite worked but uh but uh you know with various changes I was able to get them to work and sometimes it's very minor things like in his uh uh one of the things I didn't look at when I was just doing it because it worked for a while was he he put in a uh non-default network in his install config and it turned out that that was the uh OpenShift SD in one which on vSphere broke and it took me about a week to go through and sort everything out and figure out that that was exactly what broke it and it was actually just a random comment that Christian made that clued me into what the problem was you know which I discovered on Slack so it was like just total fluky luck yeah that uh I went through did that but but uh and I think one of the things that uh now I'm making this is another point in my rambling one of the things that's difficult uh that people don't always clue in on is exactly what are the assumptions that that one of these guys has so for instance on VMware uh you have to dig down deeply through a guide before you know whether or not they're requiring command line access uh which uh means that uh if you're a developer you're not going to have that uh for your company because they won't give it the the system and people won't give it to you if you have a home lab then you do okay and uh sort of in passing uh Joseph uh commented that uh you know he'd bought a uh you know VMware license etc etc etc um that's the sort of thing that you probably want to make explicit up front if somebody's going to go through a guide they don't find out at the end of it that they need this other license and then they have to go through the hassle of getting that so you know like I think that uh and I guess Charo and his guides was pretty good up front saying what his assumptions were um but that's not something you know I mean like our assumptions are transparent to us we're not always even aware of them so I think that would be sort of a useful editorial thing for somebody to look at these and say okay if you're going to go through this guide this is what you're going to need to set up or this is who it's for there are also some problems with that right um things also change all the time like with Charo I'm also discussing like maybe we don't need the bootstrap node with the next version of OpenShift anymore and so that is already a big requirement that goes away and maybe in this single node installation we still create two VMs because we need the bootstrap node then maybe we don't even need to create VMs anymore we can just install it pure bare metal without any VMs so you know the expectations change a lot and we don't know yet how they change because we are not assure uh how the development will progress there and nobody can really tell you you know the developer who's currently working on it you also cannot tell you everything that he would discover during the development so yeah right but until until it actually happens it's not real right I mean you know like there are promises but they can be decommissioned in any version going forward right like the single node cluster was at one point promise for 4.7 I think it's currently promised for 4.8 we're not there yet I will commit to the 4.8 I have heard it from above it's coming so um I saw an official red hat documentation for 4.7 I know you did but you didn't hear it from me if I oh that's different this is the dine in commitment the okd commitment um mostly to the bank take it to the bank right and um the bank of Canada and um the bank of Walmart there you go um and it's recorded here um so I'm gonna move back again so I think what I would like to ask Mike to do and maybe work with Eric um because Eric was in the homelab one is to create a template in the readme file in the okd.io much like the blog post and use Charo's homelab example because we have his blog post and retrofit Charo's into that because Charo is pretty good um at helping that and see if we can get one and and folk not not I shouldn't say homelab single node cluster approach and then also have and we're doing it have a depreciation tag too so like because we know that the single node cluster approach of Charo's is going to get depreciated so some workflow of like this is the timestamp it's version 4.8 and if I share I mean see if I can share my screen again for a second so I'm not tucking out my um here um if it's okay to think this so this has uh but didn't go down I didn't push it like it if I go to github the go down to the readme here and it may you know right now create there's a little section here create a blog post um do something similar to that um for for the deployment guide so like this is something I was going to ask like I mean we could convert the deployment guide into a github page you know that's like Jekyll head matter is what that's what's going on there so like we could convert it into some sort of like you know static site generator if you want and use git and then we could just have like you know a github pages site where all this information will be assumed in there if you want to if you want to go that route we could certainly add all that marked on there yeah um I think well for me and this is Diane creating a blog article is is wicked simple um though I still and if we know the structure of what we want and everything that's in those things we should be able to do it and I know Joseph has figured out all of the ins and outs of adding the blog into the the okd.io site so it should be just rinse and repeat what we're doing for the blog but using the structure but the blog structure is slightly it's not as hierarchical it's just yeah I guess I guess that's the question do you like so here's here's what I got so far um we want to move bare metal to homelab change the bare metal stuff to homelab um make a template uh for people to use to create their guides and I've got a bunch of information here we want in the template like you know date and version prerequisites or assumptions description hardware target audience and then some of the information we have already um and then maybe we also want to move this at some point over and so like do we want to just you know I can I can convert that to be to look more like what the blog does structure wise and use like one of these you know static site generators I'll even I mean blog is just kind of a short way to say it but really it's just a static site generator that's you know templating all the stuff together I can add that in there if you like yeah we already have that for the site itself and and Joseph went through the the so just reach out to Joseph um we're using middleman and um there I think they are they upgraded to Jekyll's a later a newer version of it so that I don't think we have to pick another one I think we can use what we use for the site um so and we can have that conversation separate from here but I think we already have what we need for the build process for the site to do it it's just um I would I think I was thinking the blog is very simplistic it's not um it doesn't have the um hierarchical structure that you have in yours um not positive but we have it it shouldn't be tough to do but no no you could certainly configure it to bring all that stuff in I guess that's kind of my question is like do you just want the files in the guides repo to kind of match the format for the for the blog and then you could just pull them in later is that would that be the best way to do this um I'm just going to stop sharing for a minute I gotta think about that um I am not a documentation specialist um so um I have an opinion um I would like it to match the style of the okd.io site um and to use the header and everything else like that but um if there's a better way of doing it and I'm going to I keep saying this is my fantasy island thing um I'll just go to porta.sh and go to yeah I'm not I'm not super familiar with porta I haven't really used it much but I mean I've built a lot of documentation sites so yeah I have not so I actually would defer to you to to make a decision about that this is sort of this is you know yeah I would never go with that style actually it's too cutesy feely for me okay what that's why I prefer I prefer I prefer the okd.io style okay thank you sorry about that that's all right I that's it's a backhanded compliment because I created it um so I mean the thing is in some ways the the guides are kind of different from even like Charo's blog and whatnot you know the guides are more just kind of data that you could look at and kind of like use as a reference um they're not necessarily formatted in a way that's meant to read like you know like Charo's posted because he obviously put a lot of time into structuring it that way um so I mean I could take some I could take some first stabs at kind of organizing this stuff and then we can see where we get to and maybe you know kind of like assess at that point like my preference at this point would just be to have the guides as plain markdown and just kind of like doing what they're supposed to do which is being a guide for how a certain deployment went and then our next operation will be to figure out like okay do you want to pull those into a website somewhere do you want to put them on the blog like let's just get the data right and get kind of the categorization right and then we can decide how we want to display it you know next you know at a later date so what I would I'm going to back myself up then backtrack so in your directory right now in the read me file um to put some best practices template information in there for people who are coming to your directory right now let's and keep and I'm paraphrasing what I think I just heard you say if you can do that so like you should be have a date time stamp or release version assumptions hardware requirements and target audience I think those would cover out a lot of things here should be in sections pick the pick whichever one you you like I'm going to lean towards single node because I know Charo has time right now and Eric was in the session and it is one that we're going to have to depreciate soon so we'll go through the whole life cycle with it so well in my head take that one and format it the way we think it should be that like as the best as the standard in your directory in your repo um and then we'll get Andrew Sullivan and Justin Pittman to do the same thing for the Libbert one and then I can tap um Joseph and um Jamie to say that again I said Charles as well yeah so that's you know that like uh but the thing is Libbert's I think actually you almost we're almost getting enough guides that you need a taxonomy or a decision tree of what you want and I think that the the number of choices is actually not that great between them so you could actually organize them them fairly easily so that would be a lot clearer um you know the sort of top decision is okay so are you are using a cloud provider or your own hardware basically uh and if using cloud provider which one then you go off on on those and it would be nice to see open stack in there somewhere I mentioned that before I'll mention that again yeah we'll get we'll get there I have an open stack and then uh the next sort of decision is as uh uh you know are you going upi or ipi if you're going upi how are you setting up the prerequisites uh you've got choices for uh ghcp dns uh uh ha proxy sort of thing um load balancing as well as trying to think of uh and then next are you using any automation tools okay so I think that the main difference between charos and andrews was the automation types of things they're doing uh charos using bare bones uh bash scripts you know fantastic you know goes way back into antiquity and still works um you know other people using terraform or you know on and on and then you get to actually doing it yeah can I ask you to write up a taxonomy just in a gist file just take that not that we're going to use that right now to frame the documentation mic because I think right now if we just got right here single node and maybe an open stack um you know that would be a good start they could you know they could be referencing other chunks of other people's documents and stuff but um but I think what you're discussing is sort of the decision the taxonomy the decision tree of which guide to use um if you're a user right yeah because you come in you know fat dumb and happy and uh I can you know refer to myself in that uh and you have you don't have a clue so where do you go yeah and you've got 20 choices yeah so the guide to the guide is what you're talking about is and then you know say so you're using a cloud provider and um you want to do um UPI and you want to use Limbert I guess you know and or a specific load balancer then then the tree would take you would have maybe a list of suggested which guide to follow and that would be like rather than trying to structure the guides based on that because I I think we're only going to have like maybe four real guides maybe you know total um and then like with home lab there were three three different you know approaches to it and we can call them approaches right um there but I think the taxonomy that you're describing in my head that's the decision tree about which guide to take a look at um you know which one fits fits you best but realize it's not using the load balancer you want right yeah with a lot of the guides you can actually swap things in and out they're not that critical to the guide it's just that somebody made a choice and it works at least this week and uh but there were other choices that were possible without making a major change to the guide I think what you're describing is a guide to the guides maybe yeah and maybe eventually it's um once we get the baseline guides in a way to restructure the guides but I I don't hesitate to um redo all of the guides as we have them now in chunks like that like yeah but that's me I and and I am not a documentation specialist I'm just trying to get the guides in the right place with enough information and expiry dates that we can move forward a little bit um and you know and right now everything just at the blog post that I'm writing is going to link to Mike's repo um and a bunch of YouTube videos which isn't super effective um and not really maintainable over the long run so um we love Mike's repo uh I I was just I totally appreciated um but I'm a little I'm a little confused at this point because like I think maybe you know some of my intentions when originally creating that those you know notes as we called them back then and maybe where this is going are out of sync here so like if I understand correctly Diane what you'd like to see eventually in those guides is material like what Charo created like a full walkthrough of how someone can do an installation is that is that what you're thinking it would be lovely if all of them were that verbose and that detailed I don't expect all of them to be that verbose but it would give them opportunity for other communities like if somebody wasn't as verbose as that they could make a pull request and add verbosity um and detail so like if someone um just put a one liner in there use you know f5's load balancer um and someone else would say or use the one that come you know this other one whatever but what I'm my goal is is it's not just verbosity but it's giving people a place to put their guides as well with some guidance about what makes a good guide and an expiry day you know so so no note not everybody is going to be as verbose as Charo and I don't want it to be a replacement for the OpenShift docs or docs.okd.io so I'm not expecting you know superhuman efforts um and as I'll go back to the beginning of this conversation one of the things that I'm trying to do is drive people to come to okd.io or content blog be it blogs or these kinds of got community guides um and but I don't know if that really helps you um but it does to some extent I mean like the the other thing though being that I have to be completely transparent about like why we created that repo and what our intentions were when we created it and this is why we called it notes originally because we really wanted to capture you know where we were addressing a problem that we're seeing internally which is that you know especially for things like vSphere um users have an incredible number of options that they can exercise when deploying their cluster and um you know as someone who is kind of on the front line of debugging a lot of these things it's really difficult for us to find good examples of how like someone like I said might have installed vMotion in their cluster and they're using it right and so what we wanted to do was we wanted to just create a place where people could put notes about like all right I'm doing a UPI vSphere installation and I'm using vMotion for all the disk images that I'm using so like here's the hardware that I had set up and like here's a couple notes on what I had to do to get vMotion working so we weren't necessarily trying to get like full like front to back like install the whole cluster guides we wanted just a place where someone could go and say well I'm thinking about doing a home lab let's say and I want to use this type of hardware what what how much will I need and they could look at one of those guides and just say well I'm going to need six machines and they're each going to have to be four core and eight gigs of ram or you know whatever someone could just look very quickly and figure out like what would they need to start doing this right um so that was kind of our original intentionality like I'm I'm I think it's interesting what you're talking about in terms of having these richer guides there that's awesome content it's just it's a lot to ask people to kind of you know create when they're coming yeah so and this may be at odds with everybody else's so the way that I vision it is rather than Craig Robinson writing a medium blog post and Charo doing wherever his is and you know and Bruce you know finding something in Kubernetes cluster laptop sysadmin blog somewhere on red hat that this would be the home for that that we would require some basic things like date timestamps of release version the assumptions hardware requirements target audience and and I just added a thought of and maybe a link to a video walkthrough of your deployment there as you know if it's available um and as much verbosity as you can um with and and that and that I think and then and then we can also when a new release comes out they this has been depreciated you know this no longer works you know or and then then we could also add in make you know maybe in the guide um the way that I think this is the way Diane thinks as an appendix of like notes about the motion you know or you know things like that but what I want to first get is the ones that we have I want to get the verbosity of that blog post in there with these minor things you know details about the date and timestamps the release version assumptions hardware target audience and any links there added in so some of them have it and stop having and not I don't want people to stop writing medium blog posts this is not what I'm saying but I'm like that drives you know that's wonderful and I'm about to do a blog post talking about those three home lab things including craigs and that and so that people drive people to watch that video and I want to be able to link back to something at the moment in Mike's repo that um has references home lab right um so that's that's where I'm coming from and nothing in your repo right now does that you have a bare metal thing and you don't have the single note thing so those those little structural things I would love to see that done in the next couple of days so that I could put a blog post out and on openshift.com and on okd.io that drives everybody to okd.io and to this and the sooner we can get it out of me Mike's repo into okd.io the more one will be able to track who's contributing as opposed to contributing to okd versus contributing to Mike's repo which is important to me as a community person um and also give credit to folks and then you know if Mike gets hit by a bus and Diane gets hit by a bus okd.io still lives on and Diane is not planning or either of those things to happen um but I also think that yeah so so my short term ask of Mike and because we get seven minutes left is to look at what you have there get the structure of the guides to reflect at least at the bare minimum what we used for the workshop right and maybe move Vadim's and Shree's and Craig's notes into the homelab one and Caro's into the single note cluster one and if you have the time um put in those um date timestamps release version metadata tags in there um and then when you get to that point maybe that's next week we look at it at the Tuesday meeting for the okd working loop or prior to that if you would like we can lift and shift it into the okd.io so it'll be running in parallel for this week and then make what I would like to do is I'll hold off on doing the post on the homelabs blog post until the Wednesday after next week's meeting to make sure everybody's happy with the move from Mike's repo to okd.io and that we've done it well um and so that's my goal by by next week to have it moved out of Mike's repo with some semblance of structure that matches the workshop um and Eric if you want to weigh in and help him with the homelab one and at the homelab with the single note cluster one um that would be great and Bruce since you were sitting through all of Caro's stuff that one fits it and Robert you are guinea pig external viewer of this to see if we got the structure right how's that that's good all right so I okay so what were you asking me to do and I'm asking you to be a reviewer a cynical okay what did you say happy and and and whatever with the other one where it was and yeah fat dumb and happy yeah there you go um depending on the day I emphasize one of those more than the other but yeah and and then you know and the ask of you is Bruce is to to really to maybe make a just a gist file for next Tuesday's meeting that is a taxonomy of um where to point people to like think of it in terms of if you know if then else if you're using a cloud provider you know yeah yeah well I mean so some of the things I'm suggesting are more long term okay so I'm not trying to get into the way of the short term so don't don't get me wrong on that you know like if we never get the short term accomplished we never get to the long term but I don't want to capture that workflow the decision tree there that you're talking about because I think it's important and even if that to start with Bruce is just a blog post on okd.io to use Joseph's mechanism for how to make a decision about which guide to use or you know what are some of that that I think is important to capture um and and I think it is a good factor in the guides yeah I will get um Charo to change that um Robert yeah I think that's just um I don't know what I don't know some of the stuff is not obsolete as quickly as you might think if you can in other words like in theory I could install 4.4 and then update it yep okay now I haven't actually tried that but uh you know because I mean things get too stale it's you can't update it anymore but it doesn't have to be in the latest and greatest as long as it still works and you get something which is updateable okay and um Robert and Eric are both um pointing out a good thing is um what license are we going to put all this under and let's bring that one up at the um meeting next week and I will um poke Charo about it between now and then about why why he chose gpl3 I don't know um and if that's a stopper but as Eric points out it's text and maybe we just need to put creative commons there somewhere um but yeah that is a conversation we should have next week so um yeah I was because I had a similar thought to Eric's when I was looking at the Apache license I was I originally just put the Apache license in there because I knew it would be yeah I think like a cc by sa or something would probably work for everything there um yeah Charo's gpl3 notwithstanding um Diane like I loud and clear I hear what you're saying I'll try to get all these changes made um you know before next Tuesday so we could review it at the meeting I think you know most of this stuff should be straightforward I'd like to change a little bit of the language in the read me and kind of update it to make it more user friendly um and I'd also like to put read me's in the sub directories so that we can help index better what people are seeing there so they're just if people come and look they have an idea what's going on um writing rewriting some of Charo's stuff is going to be like difficult um but you know maybe we can just kind of maybe if we pull out some of the information about like the requirements and whatnot we can put that in there and then leave the instructions as like a rich link back to you know the work that that he has there so we're not really like stepping on his toes yeah so what I what I really like to do is and the same with Craig Robinson is like like putting in a link a link in the template as a link to the video walkthrough I'll link to the original source so that people get cross-reference and we get some flow going back and forth like this isn't to replace a medium blog post or an open ship.com or the even though the sys admin blog post that Bruce shared with us um those things we can cross link grab grab some of it with um as long as we reference them and um yeah and and then whoever that dude was on the sys admin one he better come over and help write up his so there there you go all right yeah I'm gonna I got that I got the additional links so I'll put that I'll make sure those are in there as well okay and people are going to join in about two seconds I'm going to stop recording now