 Hello everyone. So we'll give people two more minutes to join you before we get started. In the meantime, obviously feel free to add yourself to the list of participants. Okay, so it's five minutes over so let's get started. I assume this is our last meeting before the holidays. Okay, let's bring up a new topic here. Next. So what do we have on the agenda for today? Importantly, we have been working on an initial template for project reviews. As you know, we have presented a couple of projects and some of you might already be eager to learn what's coming next and how we continue to work there. We talked to the TUC and got some examples that you also saw on the mailing list of other examples of how project reviews and due diligence is done. So when you click on that link, you should actually get access to the templates, maybe just have a brief look what's in there. Similarly, for the idea is here that obviously the team submitting the project is filling out most of this and then working together with us on the with the chairs to get this done. So it's pretty straightforward, authors are the authors of the document. So if you look to your TUC PR, which obviously should exist the presentation on the project and do the GitHub project. Then briefly discuss the goal of the project and the current status, giving an overview obviously we will learn as we go here, and also future plans so this is something that we added here so it's just state where you see the project going with the project headed. Because it makes also sense then to present how CNCF can help there. Then on the project scope. We look at okay what's the clear project definition and scope, what's the value to the CNCF ecosystem. The only reason how does it relate to other projects and also how does it want to collaborate with projects that have some overlapping, overlapping capabilities like if you for example propose a workflow engine how does this relate to other projects that are already in the CNCF are you planning to use other CNCF project. So like a simple example if you provide any type of software you, for example, planning to export some metrics data wire or where for me this open metrics, these kind of things. The alignment was under CNCF project as state okay where you already actively working with projects obviously one of the goals is here to have close collaboration with the project if you're already working with other projects. Then state here are you working with them and how to collaborate with them. We had the alignment with the application delivery reference model already in there. So some people are really included this in the project presentation. The goal here is just to maybe copy and paste what you already have, but also to define what you're expecting and what the output to other layers of the project would be or if you totally Then there's some formal requirements. He's linked to a CNCF, CNCF graduation criteria. So depending on the project we will add the graduation criteria for incubation and for a sandbox. But I recommend that you basically look at both of these criteria and just specify how you're fulfilling those criteria. So in considerations, why is this project related to cloud native for your convenience there's also the CNCF definition of cloud native in here. And then general metrics around project and code quality, which already have performance goals results, which cities is the ICD systems are using any code coverage metrics has a security assessment been done on the project and if was it. Then more work on the governance of the project. How are committed to chosen chosen how are decisions made for the decision makers. Within the project also outside of the sponsoring information so very important piece of these that we want to have as part of this review a check okay what's handled by the sponsoring organization and what's done outside of the sponsoring organization. And then the last point in there is their project independent from a specific vendors is just a vendor based open source project. I'm mailing this there were some examples like on Falco but the point to this rescue how is this project independent from what the specific vendor does in the sense that it's more than just the vendor contributing to it and grabbing the roadmap, but also not depending on some vendor related third party software. So this is pretty boilerplate from what other submissions had. So for the project to submit it until then also want to present going forward to the to see the we as the chairs want to work with you to get this document filled out with our comments, and then forward it to the to see as input for for next steps. So I hope it was helpful. Yeah. So this is Diane Mule. So is this document the responsibility of the SIG group to fill out or the submitters to fill out that just. Yeah, it's a mixed responsibility. So you will work with the SIG chairs obviously but the initial input should be coming obviously from the project. Okay, so. So the clearer you are the easier it is for us to do it. The chairs will jump in if they are uncertain about certain things or they will bring up specific questions where they're not entirely sure about this but the SIG chairs are not going to write that document so the primary step is within the submitter. So the, the, the question I have is we've already done most of this for operator framework is there an expectation we should redo it all in this format or can we move the operator framework work ahead. Now, or is it already passed or is operator framework already passed this stage from your perspective. I think that not that everything is in there but maybe we can copy and paste it over from what we have in our documents but we should definitely put it into one single spot and we have some questions specifically about operate the framework that we then also want to have as like a follow up discussion in the document. So that that's I guess what I'm trying to clarify is. So you want us to redo the, the application that we made for incubation into this format. That would be great. Yeah, because the, the, the, because you have right now that you see PR which should contain part of this information. Some other information isn't the presentation some additional information might not yet be there. So if I do this for operator framework it probably be the first guinea pig for it and can we review it at the next meeting. We can review that the next meeting and but our proposal was also that we can do it in a smaller round with the chairs beforehand because not everybody might want to listen in on all of them. Obviously we can make it publicly. Otherwise the options also that we get together once we have all that the content together there will be some questions up front. Maybe from from us from the chairs. So what I will, I will endeavor to do is to take what we have and work with Rob Samemski and Aaron and fill this out over, hopefully over the holiday and have it for you like the first week of. January for the operator framework and then I don't mind if you share it with other people because I think it's a good project with most of it is just cut and paste. Yeah, sorry for a bit of the inconvenience here but this took us some time to get started. I hope it's not too much work. I'm just pasting the link here just request access to it. I pre created the form already for. I'll do that. It actually is a good thing. So it's not a bad thing. Not that much extra work. So, thanks. We will jump in with questions along the lines and I think we have also everything nicely documented and I think especially for a specific structure discussion that document is a bit better suited than an issue or the PR. Because he is usually just goes on and on and on there's no closure. Yeah, it also lacks a bit of structure it's usually more sharing of opinion on certain things and like being related to individual bits and pieces. And yeah, you're right. You're kind of like the guinea pig. The good thing is that it worked well for other projects that have been submitted. But yeah, please, please bear with us if something is not 100% clear. We'll get it done. Hi, this is a new, you know, new to this sake, but I have a question on the reference model that we have here is there, you know, is there a requirement for projects to meet all of the topics that are mentioned the topics one two and three in the reference model, or could they be targeting a specific just one or two specific topics out of that. And are we going to clarify that. Yeah, there's no, there's no requirements to meet all of them at the reference model was really targeted as having a model where we see what have where does an application delivery project really fit in there. And there are projects within the CNCS that are very specific to one topic, but obviously you might cover multiple but you're definitely not required to fulfill all of them. If you look for example at OAM per se OAM is a way of application definition. And obviously then there's other projects coming along with it or there's also CNCS project like, for example, flux, who are more on the delivery side. But you should just know where you fit in there and what you want to consume from other layers ideally, and what you can work with but you definitely don't need to support all of them. Okay, thanks. But that's actually a good question. Okay, next topic we were also tasked by the TUC to come up with an operator definition so they and I think you will like this one as well today's a lot of talk about operators. So, the driving discussion for this really was we had this operator definition that was done quite a while ago by the chorus team. Now read it. And so what do we really want to define as an operator and they were like numerous discussions and I tried to condense some of them already into a small document which obviously contains some of my personal opinions. So please help us to get started here. This document is please provide comments on the document so overall, I think it should just help us to get this updated definition of what it is obviously we have to be as the basis we have the old, I would call it old but the chorus and credit model in there from phase one to phase five. So there were a lot of implications obviously in the technical, how they implemented technically and I kept it like very short it just really tried to take the gist out of all of these longer discussions. One topic that came out a lot was operators versus controllers. So is every controller and operator if it works based on a CD or is there a difference between the two. So discussions might be depending on personal taste to just try to sum up everything in there that kind of makes sense. What I also added in there and because I think it's summer related where operators to some extent have an overlap with what I have seen in other technologies, which is like an operator but they do kind of similar things that might be the area that's most up for discussion by others but like one thing that definitely came up with during the discussion was obviously help so it's like an opera house and operate related to help. And again for free to comment on everything here this is just my very personal opinion plus some statements that collected from other people. You're right it's get repository not get up. Related. Yeah, the other one was seen the CNEP invocation image which has like install uninstall update so it has some overlap to what some of the operator works doing. And the third one I put in here is flux and then the lower bit and pieces is pieces pieces are just for for references so feel free to throw in your comments and try to discuss them. There will be more clear statement here and we will have then a follow up discussion on this during the next meeting so take the time to read through it's not that much at your comments in here. If you think there's something substantially missing in there or misrepresented just post in comments so general guidelines for the documents as though there can be a lot of comments that eventually make it into the document and make it kind of hard to read. If you have a comment that I can or that we can work directly into the document like this example here, where I was a guitar is not correct it's really get repository. I would usually take them just in the resolve comment just post a comment back to you that added it to the document, and then resolve the comments so that we don't end up with like 400 open comments you would still see them in the history. So I hope everybody's fine with this approach, but please feel free to. Yeah, this works. This is really nice and if you go back up to the top I just to catch catch one thing. And I just added a comment so that doesn't go out. The, the maturity model has been renamed the capability model maturity had some negative comment connotations for some people in the community. So you'll see it. All the diagrams have been changed out there in the flesh. So just so people know people felt it meant that some of them were immature or whatever. But anyways, that got changed. So if you could change that that would be good so we don't continue there. If it is, I'll make this changes right after the meeting. Thanks. Good. So that that was this one. And I want to present it ideally at least the first version of the next meeting. So if you have some time over the holiday seasons, please throw in your comments. And then in the next meeting, we will then continue to discuss this one. Real quick. I really want to encourage us to start thinking of operators not a proper noun. It's just another thing. Like, like we think of controllers. And really want to applaud Coro s and then red hat for for putting this together. But for further adoption across the whole community. Let's let's turn it into a thing. So like the example, the thing that I saw here says in the maturity model that you are the capability model that we're copied out of from from red hat it says requires custom operator building simplified with SDK. We should not be pushing that second piece here. Yeah, not. We totally get that. Don't know. And that's and that's just trying to because really I want to I really want to pick put on the CNCF is not picking winners. You know, we're just really good things to the table. No, we're totally okay and we could probably even figure out how to supply you with a graphic that doesn't have that phrase in there. Yes, that was my next request because actually I'm here because it's just copy and pasted it over. Yeah, they ended you want to say something else even Brian were might might have been Rob trying to say he's he's doing two meetings at the same time. Yeah, this is Rob. Sorry, I'm doing two meetings at once. It's very advanced. I can supply you with a new updated graphic. I'm just going to paste it into the doc right now because it doesn't have all that text and it's a little bit updated looks nicer. Okay, so let's start to get them working on this one. Next one is on logo ideas. So we have some feedback. Obviously people are looking for a pack animal with spandals in different shapes and sizes. So what seems to be pretty popular are aunt and bees. Bees, bees, bees, bees. Diane votes for bees. Just to be and aren't writing a group need to ship with containers in the background might not be a logo but like the whole a whole background for us but I think overall there's like, general idea like any animal that kind of does something in that area of building something. Jimmy you mentioned you have will have some proposals in January you put it in the comments yes and I'm trying to be able to get my mic to work and my mic is not doing nice things this morning we can hear you oh cool it is all right oh there much better this nice green thing awesome um so yeah uh final call on being able to put in your ideas now uh we'll have designs to look at in January not quite sure on timing yet but uh get your stuff in this week and we will turn around towards being able to give some ideas um with the idea of being able to finalize the logo probably towards yeah February ish yeah get your ideas in now final call okay uh the next one is about the tool landscape this was a requirement we had in the beginning which led into the definition of um of the reference model and we didn't really make a lot of progress on the tool landscape the first step would I think be taking the current tools within CNCF and trying to pull them together uh into a landscape I'm really following the reference model or figuring out that the reference model might not be working perfectly for us and it says first rafter there is actually no first draft but we should be working on one and uh this is actually a call for volunteers who want to take the first draft um I want to be working on the first draft here okay everybody seems to be muted while they're saying me first please I have some cycles that I could work on this great yeah if you need any help from us uh let us know but uh we're really trying is there a way to take um to look at the CNCF landscape and add a filter that is tools that would give you a first pass for this uh in the wheel I would just anytime someone says landscape by that horde CNCF thing but it does work for some things so there might be a way to just filter on tools take out or maybe operate or add operator's tools or I don't know what you want app delivery tools what kind of tools do we mean here do you mean everything that's covered by sig app delivery or a subset of that uh I would the first step say the subset that fits into the reference model and for those tools that don't fit in the reference model we should then discuss why they're not fitting in there so is the reference model meant to um so so I'm just trying to get to to a little bit of this is the reference model supposed to fit everything under sig app delivery ideally it should fit everything in there I don't know what your take is Brian but I think that's where we should have we had it for because the main goal for the landscape was for somebody new um coming in here and looking for specific tools that provide specific uh capabilities they can look okay I want to ship my application I don't need something I don't know for application definition what can I use or I need something for uh deployment what can I use and then have a look at what these tools do and then ideally can you can link to the individual uh well actually to the project and then link to those projects so then is looking at all the projects that would fall under sig app delivery and putting them against the reference model kind of a way to test input to the reference model to see what works what doesn't and maybe exactly yeah yeah I think it does test the reference model it might also help us define where some blind spots are and where we may be on the over and uh on the representative Amy is there an easy list easy way to get the list of all cncf projects I think you can pick them from the website but you would have to like really go through especially with the sandbox no actually there is a way to be able to sort by sandbox and there is a way to be able to sort by tooling right now um it's under groupings and there's cncf relation ah okay so on the cncf website uh yeah the cloud native landscape I'm poking at it now to make sure but yeah you can start by cncf relation um it's called interactive that's obviously we won't do the whole thing yet yeah uh here we go there is a repo where you can file issues to correct any like things um uh or like you know requests um and it's under the cncf landscape uh repo uh it just seems to be not working at give it a second to be able to load it takes a little ah okay it takes a second so it takes longer than you think it should so this is just a static page it's a web app that loads all the information for all of this in the background and then all the filtering and everything happens in browser so you're pulling in over a thousand different projects and all their metadata in and then your browser can go and cycle through it so it does take a minute sometimes to load okay just download the internet okay that's fine I think you might need to switch to the card mode for some of the filters to yeah and then on the left yeah just this this is Diane I I brought up the the topic of this thing and and it's not my favorite thing but the source code for it is um out there in the real world in the some github repo and I repurposed it for like uh an an mi an uh an ai ml landscape um here at red hat so the same look and feel so we you know whether we use an add a metadata tag to this landscape for app sigg delivery reference model for tooling whatever that we want to call it I would say we should do um it's not hard to do it here because this is but um or we could just stand up our own I just don't want to reinvent the wheel but I totally agree if you can yeah definitely have a look at the source code maybe we can repurpose it for our layers there is documentation for being able to set up your own as well just wanted to post a proper link I won't do it today obviously uh so once yeah once this is done I think I'm going to post this one and just put in here that we are going to the easiest thing would be for us to just add the metadata tag to the existing one so we don't have to maintain it yeah I agree I'll have a look at the repo and uh what can be potentially done there and I'll post in here a link with the settings that I think which I approach it we should look at so feel free to uh revisit this once this whole thing has loaded here good uh yeah let's put this in on the agenda for the next one next topic is continued work on air gap environments I think we had a great presentation in I think we've done on this on the tool landscape so far or are there any other comments on this one I think we have action items and should be good can you still hear me again okay next topic would then be continued work on air gap environment so we had this great presentation the last time on best practices for for air gap environments it was during cube kind of a lot of interest on like how to set up and deploy to an air gap environment and Jeremy you referenced also the CNAP documentation on this one uh I think we have something there that's already quite good but potentially could invest a bit more work into this if there's a proper um yeah if there's proper interest from the community so this for me is more or less a talk to who else in the community is interested there was a lot of interest at kubecom but before putting into more work for a white paper I would wait for other people who want you to contribute there instead of putting even more work out there yeah I would agree I think um it'd be useful to maybe put out a message on the mailing list or something to solicit any interest and maybe we can do a quick survey to see if there's any other projects that are addressing that somebody replied to the the the recording on youtube with a link to something that was kubernetes specific and I think somebody mentioned that there's something in the operator world that's kind of addressing this as well um somebody at kubecon showed me a helm plugin that did the same thing so it'd be cool to maybe survey those and solicit those people to see if they would like to contribute as well before we do yeah I agree so I think maybe stop to distributors like SAP or IBM may be interested in this topic because if you have the problem how to deliver application or air-gapped environment I think we need to send the message to the to the outside world to let people know hey we are talking about this topic so if you're interested please come if folks just wanted to throw it out there that um red hat and open shift have done a lot of work on offline stuff for operators and the life cycle manager cool yeah I think I'd be ideally if you continue this send out to the mailing list to wants to actively contribute to this topic just reply there and then we can create like a smaller working group around this to to move this to the next level great that now people are supposing to youtube links um so even more channels to find information maybe uh as I don't know what we refer people back on the youtube can we just can we just add some have to look at whether we can add this somewhere to the youtube channel at the best posted back to um to the github repo as an issue or they want to have commented to the mailing list because otherwise these things might simply get lost and we want to ensure that we uh that people get hurt about this one because I honestly did not look at youtube good so yeah we'll set this out by the mailing list ask other projects that contribute there and also the bigger software vendors uh who's interested and obviously other providers uh next topic is the next um meeting as the holiday season is coming up I just wanted to clarify when we want to have the next meeting that's the next oh we already canceled the January 1st one which I think is a smart decision yeah we'll be the 15th 15th okay great yeah everybody will be back yeah no one's coming on the first I hope not even better things and talk about software hopefully yeah yeah any other topics anybody wants to discuss or things they want to bring up otherwise I think we are pretty much done okay well happy holidays everyone happy holidays yeah see you in the next decade yeah crazy all right yeah all right goodbye