 Hello folks, we're waiting a few more minutes for a few more people to join Sorry for the delay folks for cheese computers acting up. He'll be on shortly To join us And welcome everyone. It's our first It's our first time. Okay. Well, we're while we're waiting for the last few to join I would like to say welcome everyone to our first actual sick observability meeting As a sick as a newly formed sick This is a CNCF event and the code of contact applies So everyone be yourselves, and I'm sure it'll be great Amy let me know in the future if there's an actual script. I'm supposed to read there. Nope. You are good Okay Let's let's give let's give Richie one more second. He's he's gonna be online shortly All right, so I hope rich rich will be able to join us Why don't we go around the room and give a little quick intro? I know I haven't met some of the folks here And Steve just joined as well So maybe we could start with him Last one in first one out Or first first one up rather Sounds great. Hey folks. My name is Steve Flanders Director of engineering at Splunk. I am an open telemetry Collector approver and I've kind of been in the observability space here for just under a decade at this point working in the logging space at VMware I was at a sort of called omniscient and we helped co-found the open census service and now working on open telemetry for Metrics traces and we just recently started to log sick. So logging coming up as well to really make it easy to get Floatry data out For both applications as well as kind of an agent and collector and super excited to be here today I guess we don't have a consistent list in order. So I'll just go from the top as I see him Jonah yeah, Jonah Cowell here CTO over at logs IO and Just getting us involved in what's going on around cncf and some of the standards and We're looking forward to contributing and participating. So thanks I'm currently busy trying to let a whole bunch of other people into this room. So uh, good to see you all. I'll continue that work Is it my turn Bartek? Yeah Cool. I heard something. So, hey, hello. My name is Bartek. I am principal software engineer at thread hat I'm prompt use maintainer and co-author of channels And yeah, I was involving observability for a long time as well and super happy that this is started. So, yeah Hello, city leclerc. I work for uh elastic where I am a product manager working on observability on Exactly to be there as well Well, you know, I why don't you guys can just uh, we're gonna be working a lot together Uh, don't make me redouble us just everybody say hi and we can tell sort I'll go first. My name is matt First of the unsorted. Uh, my name is matt young. Uh, I'm uh, bb of power engineering at everquote Where we're running uh communities and a bunch of cncf projects across a couple different clouds Yep Let me go next i'm richard Uh, but we'll go for working for gulfana prometheus team and uh, the second chair together with matt I can go next. Uh, i'm galkin I'm a prometheus and cortex maintainer and i've been Involved with prometheus hence observability for Nearly four years now and I work at rafana Hey, I can go next. Um, so i'm a tias. I work with bar take on Like at redhead on Thanos and meteos and kubernetes and whatever kind of observability and goke Things we need to do Sorry, go ahead. Uh for me Yeah, hi folks. I am pranay. Uh, I am a co-founder at signals. Uh, you're based out of banglore india We are a pretty young startup working primarily on monitoring and Observability, so happy to be here participate and learn from you guys. Thank you. I guess I can go next I'm kamal. I also work with mathias and bartek at redhead observability platform team Yeah, I am also a contributor to the tana's project Anyone else? I just joined morgan mclean from google Hey, hi morgan. Hey Uh, did we hear from thomas yet as well? I noticed he's here He joined us. No, sorry. I'm here. I'm uh, Engineer over at buoyant working on uh, linkerd for service meshes I'm irish kuro from uber, uh, maintainer of yeager open to the metric I don't know if we got everybody or if anybody else wants to go I'll give it 10 more seconds and then we'll we'll get rolling here three two one ten, okay Cool, uh, so we have uh, again, this is our very first time meeting as a sig So i'm sure we'll have bumps and starts. We have More potential work streams and things to do Then is even Able to be talked about in one quick meeting So i'm very excited by that as well as excited that we've got a sizable turnout And a lot of interest from both open source and industry Participants Does anyone see anything on the agenda that we think should be there? That's not or is there any kind of hot things and feel free to just Write stuff into the agenda again. This is meeting number one. The process is emerging And I want to make sure that we're not too prescriptive about what we do or don't talk about for this this first meeting There's nothing So there's the aob at the end and I left in some some time on purpose But I think we're quicker with the janitorium and stuff. Anyway, uh quick update on current toc status We send email to the toc list on both Getting the third chair approved as the flanners who's also on the call. I think Yep, I'm here. Yes, there is And also getting Bartek as the tech lead I don't think I saw any answers to either email threat yet Yeah, no like I don't know um I'm going to poke cncf uh or the toc about about actual votes on this course obviously Um, that's that's false on toc to um to either ask questions or to vote I hope that we will get questions and or votes soon Of course, that would be better than not telling them So, uh, please add that to the uh capability kind of update in there. Thank you Sorry, I couldn't really hear you Uh next today should be the next toc meeting Okay, and we should be adding this to the agenda Yes, you have a slide on the agenda Perfect, then I will write that down and take this as it to do or actually let's Let's check to do in the document I'm just going to do it in prompt and we can figure a code way out after Okay, so that's basically the update of on the toc. There's nothing Nothing else to be to be said currently um Overview over initial work packages. We have identified two work packages for now One is cortex incubation and one is Thanos incubation um, of course those are kind of Currently in a in a place to be to be evaluated Cortex in particular they they did all the steps which need to be done for review if People are in this document if you click the github link I can also screen share if you prefer but you can just click that github link there is a Well a process, uh, which part they copied in basically the sake needs to to make a recommendation And then the toc will be reviewing that recommendation and and decide on if they want to If they want to proceed with incubation for that project Yes or no, and then there is more there's due diligence. That's not clearly defined Who will be doing it? I suspect that part of this will be falling Back on us and part of it will not be and then Basically, that's the way we'll be going So, um, I know Bartek has some thoughts around what he wants to do in the cortex review So maybe Bartek just says a few words about what he thinks he would be doing Sure. Thanks, Rigi. So, um So essentially we are pretty close with the cortex team, right? So we were talking about this kind of graduation process For both Thanos and Cortex because like these are pretty similar And essentially what we need to do as a as a seek observability is we need to provide some kind of recommendation Based on the graduation criteria From sandbox to incubation, right? Um, and there was no clear form and like template for such such thing But our idea is to create something that will kind of Fit into the due diligence document template And then this template is like well written and there was like formal kind of template for that So, um, this actually answers, you know, those criteria graduation criteria, but also It would be just easier for to see to create like official due diligence We hope so, right? So, um, I think that would be the idea to just focus on the on the for Cortex first and then maybe for Thanos We thought about sharing those kind of because the project is super super similar, but like After some discussion, it's not clean. That's not doable because the questions are really personal for project. Let's say like what customers you have. What's your usage, what's your kind of commuters kind of base looks like so Um, ideally it will be kind of similar, but again, we need a proper document for each project. So, yeah, that's the first step and one kind of idea from from my side is that it would be nice to get also The maintainers of those projects to to build this document with you together or like with us, right? Um, so we can collaborate on this and just have it quicker, I guess um, so Like as one of the maintainers of Cortex, uh First thanks for taking this up and second, how can we help you like should we Fill out the due diligence review template and share it with you folks Yeah, I think that would be that would be the way to go Okay. Yeah, I think that was the most helpful when we did Yeager presentation to TSE It's like maintainers know the best how to answer all those questions and then we can take it from there Yeah, in addition, I was going to mention the Yeager Both the Google doc and then the resulting markdown, which is a little more Shorter are both. I think a pretty good example of due diligence performed effectively and deeply I think the the interview is a good idea On the other hand, we should be mindful of of the meeting time Which we have and try to do actual work also outside of just a meeting So, uh, so writing things down both questions and answers and and assessments and everything Under the week in my opinion is even more important than having a interactive Q&A On the other hand having the Q&A sounds really like a good idea and we should probably schedule this for a next call I've also put in links over into the chat as well linking over to both the graduation criteria Which I know there was a question about as well as the due diligence And the last link is the preload chart for how this actually Works. So this is part of both the SIG group as well as being able to get a toc sponsor Correct. Yes. Awesome. That's the that's what um Um, uh was linked in the github issue as well on the same chart. So Cool So, uh, gartham, do you want to take an action item of making sure that we have a critical mass of cortex maintainers In the next call for any questions? Okay. Yep And for everyone else be more than Feel more than invited to to put questions We can either put them in in slack or on the mailing list or if people prefer I can also just create a new document Where we can yeah, maybe let's do the document right now. Any any weight those towards the document? um So I would suggest that we do this after the like after bar text performs the evaluation And then shares the results I think the questions can run the can run largely in parallel, but I don't have a strong opinion both is fine in my opinion Yeah, definitely can can be in between I think I think it might also make sense as part of all of this for both cortex and thanos and anything else that the SIG is going to provide assistance with the due diligence on Would be to engage with the community and send out a survey to see who's actually using cortex and thanos Or thanos or both or or whatever in in production or in staging Perhaps the end user community would be a good place to start Just to kind of get just to make sure we're not missing any people that are already Within the umbrella of the cncf and have direct experience with these that could lend insights to that That might also be a preparation or or input to the toc course they need to do user interviews anyway Amy, do you have any thoughts on if we would be preempting or or Doing the work of the toc differently, but slightly wrong or would this actually be helpful? Amy did you I think she is not here. Yes Yeah, she I'm going to poke her on I'm going to pop currents like I I just want to avoid that we end up in a situation where basically do The same work subtly wrong and then toc has double the work or the user's complainers. They have to do it twice Just The question from my side, let's say we have this evaluation doc Let's say we gathered all the input from from the cortex team How we agree as a As a kind of observability seek group That our I don't know review is is correct or not Is the voice of the seek not only I don't know maintainers and me or things like that Call for consensus I I strongly believe we should be following itf rules on how to how to attain rough consensus Um, of course, that's that's been shown to to work reliably in in pretty much all situations Um, and if the sick as a whole finds rough consensus, that is something to take and bring to toc But that's not the chair describing or prescribing for the whole Group how we should be doing it. That is an opinion which is open for debate So anyone who has different opinion now is the time to voice it Of course We will be starting to set the tone for the whole sake over the next few calls Makes sense to me any objections That sounds good. Yeah, I assume we're talking about is rfc 72 82 On consensus and humming in the idea So Yes, I'm sorry. Yeah, it sounds great that there's actually an rfc on this Yeah, there it is. Um, I've been referring to this I don't know how often I try and structure much of of online work based on the itf guidelines because They have been better proof Yeah, okay, so um, then that is probably the first consensus of this group Then let's play then that's actually something Um, if you can look at the document for a second, that's how I usually do it in the premises team I call it the game of consensus So now I have this written out. I'm just looking for the rfc so we can actually Link the correct one And the way which I usually do it is I mark it. Oh, thank you Ask the whole group to please read it And either speak up or you have agreed to it and only if there is consensus, then it's marked green Uh, also this kind of uh implies that the Attendee list should be should be complete Of course, obviously the people who are on this call have now agreed on Whatever there is consensus on or we note if there is no consensus like if there is Someone who strongly disagrees we just write below it person x y things We should be doing deathmatch in person instead and we just note this down to to give the the dissenting voice an actual voice and not just ignore it and And minutes of meeting that that dissenting voice away So is there a consensus on this one? three two one Very good While this call is ongoing you can still voice concerns After you can't So, um as far as the uh six structure we uh supposed to have chairs and uh tech leads, right? Um, I I don't remember the charter Exactly does charter give any specific rights to just members who are not chairs or tech leads Because if not then the other option to solve in this is simply a vote From the tech lead on the email thread. Uh, this is how to see votes. And so we could do the same thing And so yes, it's it's still like consensus because if someone objects then people need to I guess um Result that or you can even do simple majority vote In my opinion, um The the the role of chair is very much one of facilitating and and finding consensus and feeling possible Consensus if there are disagreements, it's very much not Overriding the group Also, I would prefer not to have votes on things where we don't need to have votes on Of course, a there are Harder to override or to adapt later when when new facts emerge or something and also they're just more hostile Overall, it The the atmosphere of the group tends to be better if we are doing consensus based decision making as opposed to vote based decision making That being said for example, we can if need be introduce a mechanism where We vote on something. Uh, for example having a recommendation to the toc But I would prefer not to we can also have a system there met and me on potentially steve or hopefully steve Will uh, will approach uh toc and say this is an actual decision by the chairs But again, at least personally, I wouldn't prefer to do it this way I would much prefer to to actually work based on rough consensus as per itf But again, this is not me as the chair saying this is how it needs to be done This is me as a member of the six saying I think this is this should be done this way Yeah, um, Amy we don't have Amy here to to really give the official voice of the toc but I am hopeful and I and i'm optimistic that in particular for things like Do diligence evaluations of projects that want to graduate from sandbox to incubation? That we do this, you know sort of together With a consensus based approach And you know, I think if we have like a lot of division around just the fundamentals of Our requirements being met or is a project stable or meeting meeting the criteria to move From one phase to another that these aren't controversial And you know, I I agree. I don't want to spend too much time being led to just upfront if we can just Work together in an open way Where we're inclusive And we're and we're being careful to over communicate and welcome You know various viewpoints and and out of that plurality of of views and experiences, you know, we can have A true consensus from the SIG I think that Also, it's my understanding that as a SIG we're free to make our own rules that work best for us and the The the role of a chair as prescribed by the toc is really as Richard said to facilitate not dictate You know, we're really here to to foster a community and let that community self-govern and self-organize Um, does that help uh, Yuri answer the question or yeah And and also if folks have, you know, this is our first SIG. It's my first SIG involvement with the CNCF At least so if there are other folks from other SIGs that have feedback or insight into As we move forward, please Be vocal Just as an update in between Amy asked me to suggest User interviews to to see next week, which I already put as a to-do for myself And I already asked her if it's fine to run on consensus. She's currently putting in a different meeting, but we'll get back to us hopefully within that Within this call, I mean I had a question What's what's the role of the toc liaison for the SIG? Are very clear I'm not 100% clear either, but at least part of the responsibility and part of the role is to To basically be our our interface towards toc Uh, which obviously is also In part being overlapped by us simply being part of the toc calls and and being able to to talk for ourselves I would say that person is most likely the default sponsor for anything where we say something should go for incubation something should go for For actual graduation, I think though that person is then the default sponsor Also, maybe a person to out navigate things if things need to be navigated that's at least my rough understanding, but It's not completely spelled out that I could find but I might be wrong Yeah, um, and I had a fairly long conversation with brendan our liaison brendan burns He he has a lot of deep experience in cncf projects and Your I'm so sorry. Um, it's still getting used to zoom and and working in isolation both I had a long conversation with brendan burns Um, uh, you know to toc to secure his involvement as our as our liaison He's an advocate for this sig at toc So he's in you know, if we need something that we're not getting If we need either funding or logistical support or if we're orchestrating things like toc votes On you know, once some of these due diligence things are done or if we're making other recommendations he really is the Our advocate and our proxy in that voting capacity at the toc So and again, I'm sure I'm butchering and I'm sure that's not a complete definition But as I understand it, that's roughly his role His role, however, is not to be Um, I mean he I welcome his involvement. I hope he joins us in some of these things I think he has good perspective But um, he really is there to advocate for us not Do things necessarily with us at least as prescribed by that role if that makes sense It does as an update I got the next answer by avie. This is actually working pretty well In the charter operating model as a starting point Let's be inspired by cnc f os s projects and by kubernetes six. This means minimal viable governance and community-based organization Looking at prometheus That's Pretty much everything based on consensus As far as we can possibly make it for obvious reasons So, yeah, I would say we are completely free to define whatever we want to do with the explicit recommendation To be as lightweight as we can possibly be so I would say It's well within our rights and also the intention of cnc f For this sick to be consensus-based So that being said I think that consensus Can stand as it is that we follow it ff consensus rules and if it turns out that we can't we will just change it Like if we if we if we tend up or if we end up fighting all the time We will have a world-based system anyway, of course that would be the only thing which would get it all get us out of that home But until that time and that time is hopefully never Let's do it lightweight and consensus-based objections Okay, we diverged massively from from uh cortex the only open question which I still have is Should we have a rolling document for questions updates? Whatever regarding the actual evaluation of Of a thing and let's let's cast this wider than just cortex Should we have one single document where we collate everything? Which is related to review of that thing? Or should we be doing this in github issues or should we be doing this in in our normal meeting notes? I wouldn't suggest you do that, but just to have the option. What what do people think? rolling document issues Polar requests markdown I guess you mean in some kind of questions or like suggestions for the next Um evaluation or things like that. What was useful? What was not right? No more like ongoing stuff. So um, I'm I'm somewhat worried of Questions being overlapped or dropped if they're asked on slack or on irc or on the mailing list or in the call Or meeting in person once we are able to do this in two years time or whatever And basically having one project or one review or one workflow specific default document Um From my side, I mean github issues make sense for like a synchronous um reporting, but Maybe it's too too too heavy. I'm sure what people think If you expect discussions then google docs are better because there's a they allow more fine grained threads And so github issues become pretty heavy in that regard We found that when forming the charter as well. We started with a github doc and just iterated there and then Moved it to markdown once it had solidified and sort of a rough consensus Yeah, that's actually a good point. I think that part that document should then become part of the permanent record And I think if we do a pdf export of the document Then we can also export all comments Which would basically allow us to to to snapshot the whole discussion And then just make that part of the good repository and be done with it Which I mean also linked back to the document obviously course course then you have the live source But for for prosperity it probably makes sense to just drop the pdf into the git and Be done with it like to the document that that thing Should we try it this way or any other any other suggestions? the only question And the only question I have is if you are doing google doc, how do you prevent random editing as in Like if it's publicly editable if anybody couldn't edit We have the timeline if there is abuse We can either lock it down to name the counts or we can find the the person who did it if it was a named account and Have a conversation You know what what we did with the charter document too is we enabled it for Comments and for like suggestions and then you know initially It wasn't a problem that way people can you know make suggestions and we can review them You know, I don't want to be too prescriptive, but We have a fairly small set at at the moment and I'm guessing we're all gonna Get along well Yeah, I think that's start with editing right and The trust is high and so if there is issues then we can switch to just the comments only Well, I would also tend towards the same Of course it lowers the power of entry to newcomers and that is something we should actively strive for in my opinion at least I don't really believe in solving technical problems. Sorry social problems with technical solutions unless It's really about locking out So, you know that that sounds good. I mean we have morgan here. We can just ask him to deliver until next call We're only locked in users can can edit a document which is On open telemetry we we ended up for some of them ended up switching to comment only not because of like People being individual people in the community being bad actors But there were some where we would just get literal spam Like literal like like advertisements getting pasted into documents by isoom bots Some of them some of them we ended up switching to comment only though. I think many of the meeting notes are Are publicly edible As long as the link isn't like if the link is just embedded in the calendar invite I doubt too many bots are gonna find it. It's it's more if the link is posted on a webpage somewhere that Something will probably happen eventually, but we always have the edit history Yeah, we can always change it it's not big deal Yeah, truth be told when we did the charter stuff I would after before and after every meeting I made private copies of it just because I wasn't sure if there would be a bot attack one day Yeah, I think I think if other people edit edit access they can't change the history So I wouldn't worry about it too much We decided to open the edits and if it's a problem then we set it to comments only and explicitly invite people I had one or two instances where I actually needed to troll through Through the history for reasons And it worked fine and in one or two cases. I actually had to revert stuff and it was super easy Or anything else that being said for the public we had this for the public Prometheus ecosystem call recently were exactly by the time quarantine started We got trolled by by people joining zoom calls And then we had to revert the the public doc and restrict it and it also worked without any problems So I wouldn't worry too much No, but this is this is um Actually correlate. I mean we are diverging from the initial topic, but this is actually highly Correlated when I was still doing staffing for free note. I actually had all the school holidays wrote white in my calendar simply to know when the trolling would take up a then course all the Adolescent people stuck at home with nothing to do and that's the same for the lockdown But those this has quieted down quite a bit as far as I can see. So yeah, let's let's start open and we can always lock it down more if needed So do we let's Game of consensus, I would say let's Keep it What why are typing I can't remember we actually just talk about it, but I would assume that After we've kind of reached consensus in that sort of document then there is a form a more formal markdown in a pr So that kithub becomes And the sick observer really repop becomes the ultimate source of truth So that's we can we can easily copy that stuff over. That's not something which I've done previously, but it makes sense in this case because this is more rolling um, yeah But yeah, that's that's actually a good point We can even maybe have a consensus dot md where we just track consensus items Back to where it was made. So you have a short overview with also where the source lives That should actually be quite lightweight to do on the side Yeah, I like it So I'm marking this in the document. I hope everyone is in the document. I can also read it out Suggesting for consensus. We have one rolling document of our work package This document is as open as possible and we can lock it down if there is abuse All agreed good here Sorry, what good here agreed good Objections very good Yeah, AOP is any other business Hey, I'll be vulnerable and transparent and say I didn't know what it meant I used to hate the term and some at some point that just became part of my Yeah, I I was a stinker and I populated any other business with a couple of things that um, kind of prepared a little bit better. Um this morning So Should is there anything more for cortex? I think not of course we spend ages on cortex Okay, you're not a bunch of stuff along the way. That's cool Yeah, no, I'm I'm like and by and large at least from my perspective having this kind of ambling thing Where we sometimes switch focus and just talk about more stuff as long as we get shit done is totally fine Um, it's it's mainly about getting shit done, but like touching other topics can sometimes be super useful So Thanos It's totally the same discussion as with cortex. The question only is about the you know, how When to to kind of touch it Should we do it can we do it in the same time why we're all kind of switching context and blocked on Nothing should we totally don't touch it before Yeah This one I actually wanted to throw a question out I mean to to all of you Thanos and cortex in many cases have contributors that talk quite a lot There's some code sharing going on which is healthy Between the two projects. So do we want to treat the due diligence for cortex and Thanos? together holistically Or do we want to make them separate? potentially parallel work streams I could see both being valid. What do you all think? In my opinion, they should be two separate things, but obviously there would be tons of overlap So I think we should be encouraging to just basically copy stuff over where it makes sense But we should end up with two distinct recommendations for to see But that's my opinion not my statement. Like what do I think? Yeah, from my side to this thing thief to separate things makes sense. The only question is like Can we have those two distinct with things in the same time or Doesn't time it matters, right? From I think there is two two possible answers one Especially at the beginning we shouldn't be overloading the sick with too much work Because we don't know how much work we actually get done So we should be more optimizing towards getting stuff done and not opening new new new stuff On the other hand, especially with cortex and Thanos, it probably makes sense to treat them More or less at the same time course a lot of things Will actually be the same for the two and also a lot of things will be Will obviously be overlapping with Prometheus and you can just refer to okay That's how Prometheus did it and see over there and they already graduated So I think it's somewhere in between But again, what do others think? Got down any any feelings from the cortex side? Got down is even here Probably dropped Let me poke him I think he'll agree. I mean, I'm not him, but I think he will large he won't object at least Yeah, I spoke to Gotham and Tom offline Tom Wilkie and they were both into into Waiting on this Thanos side as well So yeah, it looks like everything is in an agreed direction. I think So as we actually seem to have agreement on this Let's do the dance of consensus or the game of consensus again So propose consensus Cortex and Thanos should be treated as two separate work packages and you're in result in two two separate Okay, let's start again. Cortex and Thanos should be treated as two separate work packages And you result in two separate proposals for to see but we should leverage the overlaps objections Sounds good. Okay. Any other business Should we just move into what was written below or I think yes, of course, that's the order of whoever came up with stuff Okay, work stream ID create CNCF observability roadmap. Um, who put that in who wants to talk about it? Uh, I did and I just realized I have I might muted by people so I'm typing sorry um, yeah, one of the things we talked about during the charter was just having an initial Graphic or something that's just a subset of the overall CNCF roadmap for CNCF projects related to observability Particularly since there's a lot of different projects in overlap. There are other SIGs So this is sort of a initial site content to to provide a lexicon and or a pictorial Viewing of all the CNCF projects related to observability perhaps we I would propose we just create some github issues for these and people that are interested in working on them can Start and can self organize. We can use the SIG of your ability channel I don't know if we're gonna lose them. We're almost out of time here But that was one idea most of the stuff that I put steve flanders When I talked to him a week or so ago Had a cool idea as well I don't know. I don't want to speak for you steve Yeah, I can talk about it real quick. So basically the idea is to kind of show off some of the observability projects In a more concise way where you can kind of get a feel for what's out there. What's available how you can stitch things together How you can like get started versus like get to a production environment I think it's a good way for Project teams to even collaborate with one another and it also gives an opportunity for everyone's kind of opinions and thoughts to be shared It's pretty common that you'll see like experts in the field kind of talk about their particular area Maybe on medium or on their own personal blog But being able to bring in a kind of a more diverse Community of people that are kind of in the observability space and some of the problems they're trying to solve Be that blog posts videos training demos presentations. What have you I think would be extremely valuable just having a constant stream of of information that's being provided that people could subscribe to and Just learn more about what's possible in the observability space constant stream or reference Where where people can look up specific user stories because what you just said would fit both but Is it both or is it one or is the other? um, I think the it's kind of both the idea would be that you could go to a place where there is information That's being released. It's not like a one-time dump and then it's like Relevant three six months down the line But the idea is that it's constantly updated like conferences happen throughout the year They're of course virtual now blog posts typically happen on some sort of regular cadence at least that's ideal instead of just like one time And then nothing for months But some of this information may be reusable so like videos Maybe you don't have to like re-record the same thing over and over again unless things change But I think the idea would be that there should be some sort of cadence where like materials being released pretty regularly That way you can get a feel for things and if it grows big enough this could even be different categories Uh, so maybe it focuses on a subset of observability or maybe it focuses on like stitching things together So kind of open-ended to start right i'm kind of curious what the community of the whole thinks what would be possible But just bringing some of this material together and starting to show off some of these projects I think would be extremely valuable The the one con I agree. I think it's a good idea The one concern which I have recently in Kubernetes There was a thing where it was more or less like advertising content on official Kubernetes channel Which led to some fallout. So we should have some some Some standard maybe some some written down standard of of bare minimum of what Technical level of proficiency you need to actually get into this distribution Because if it's going to be abused as a sales channel or as a marketing channel, we will all have a bad time Other than that, I think we can just Go ahead and try it. Yep. No, I agree. I definitely need to be some sort of like contributing guidelines here What the expectations are this is not meant to be a sales thing It's the goal is to promote and show off the observability products in cncf and how you can get them together But I totally agree. There needs to be some guidelines in place Okay anyone else It's it's it seems as if it's always the same people talking It's the same on me me I mean Basically, I agree and I also am very like slightly concerned with what you just said Like there were a couple of blog posts all over the place in the like last two weeks So I definitely see that and I think I don't know like I guess we we could have something where people kind of like can propose blog posts or whatever Uh resources they have and then at least like Some some people how how many that these are I don't know but like someone From from the sick should look at them and at least like say Like hey, this this seems somewhat reasonable. Where's we've seen things. They were just bluntly wrong sadly Which is something that we shouldn't promote Right Like otherwise, I'm super happy like like promoting all things observability. I mean, that's like literally What we're here for right Cool, like I guess Steve do you want to take a stab at maybe Yeah Trapped in some of those guidelines and then we can bring it back to the SIG for consensus Yeah, let me let me start a doc on this And then we can socialize around and get people's feedback Um I think we have what four more minutes Um The third one is really every statement of the first one. Uh, so the the last thing, uh I wanted to make sure we at least threw out there to think about for next time is Um in our charter we had called out a lot of stuff. We've already talked about Um, I think webinars and presentations, uh, is actually a little bit Even more well-formed than what Steve was just talking about. I'm if I understand these are These are directly from the project contributors or the projects themselves to actually do a more deep dive webinar so perhaps Generating a backlog of those is something that we could do initially And then I was curious if there were any folks on the call today We can talk in the slack channel as well between now and the next meeting But uh working groups or something we can opt to form. We don't have to Um, and I put a link to The charter and then the charter. There's a link to the toc's definition of a working group But does anyone Want to start a working group on any particular subtopic or subdomain? You mean start Making webinars and such or uh, no, I'm sorry in the interest of time. I just skipped along to the third bullet there So as a sig we can opt to to form working groups if if that is Like if there's some Okay groups are meant to be time bounded. They're meant to be like here For example, we could as a sig just say hey, we should have an observability roadmap And someone passionate about this can like make a cool graphic and Submit it as a pr and we can be done Or if we wanted to make a more in-depth longer report that says here is the current state And we should probably do this at some point Here's a current state of all the projects and the what we expect to come And like when they might want to graduate to the next phase and it's a much more in-depth You know accounting that might be something that a working group does and they you know They form for a month or two or whatever it is But in some time bounded way with it with a the miserable outcome And then they can go have their own meetings and you know They're part of the sig and that working group can then bring it back to the ultimate to to the larger body I don't know if we have enough people But I just want to make sure that if anybody had come today It said I want to go start a working group because I know what they are and I think we should do it I wanted to make sure that they had some airtime In our first meeting that's all Got feeling to be honest. I'm somewhat worried of starting all the things at once I would much prefer to see outcomes like the time boundedness and the measurable outcome. I like very much I'm not convinced. We are at the stage as a newly formed group where we can already start splitting out without basically relying On this kind of cadence and and synchronization mechanism of the call of the channel But that's not me saying no, we shouldn't I'm just worried about doing this day one I I concur and again to be clear. I'm not proposing that we form working groups now But again, I want to make sure if somebody showed up to our very first meeting And that's what they really wanted to go do after reading our charter that we we let them Have a voice it sounds like Sounds like that's not the case. So Is that inverse consensus? silence is a I think we can establish silence as agreement to whatever was said because else With 15 or even more people hopefully over time it will become really Okay We're actually over time. I I should already be in the next call um Yeah, I think this call was like scheduled until 10 uh, 10 minutes before The hour Right And then there were two Yeah, I actually Wanted to leave but I didn't I'll catch up later. See you. See you. Thank you. Bye