Kubernetes Steering Committee, 3 Dec 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Kubernetes Public Steering Committee Monthly Meeting for 20211203

Description

Kubernetes Public Steering Committee Monthly Meeting for 20211203

A

All right, we are officially recording hello. Everyone welcome to the last meeting, oh black public meeting of 2021 into 2022 already. I can't believe this year is behind us at this point. uh This is the kubernetes steering committee meeting. So if you are in the right spot for governance, hello, if you're not go ahead and click off, um we do have a code of conduct today, please be excellent to each other. If anybody does have any uh issues, please feel free to dm me or email conduct, kubernetes dot io.

A

uh Also, if we have a notetaker, I will also be taking notes too, but if there's anybody on the line that would like to help us take notes, that would be most wonderful.

A

And I'm also going to be keeping an eye out for zoom chat and let's get right into it. So first things first on the agenda. We do have a very jam-packed agenda today, so I apologize to any community members here that are coming in to add something to the agenda.

A

uh We are now full, but please feel free to add any agenda items to the very end of it, so that we can keep track for our next meeting please. I definitely would still like to encourage folks to add things to the agenda, even though it's full um and so first things. First, we do have a really quick vote that we need to get down and add our lgtms into uh the issue.

A

This is the issue for liaisons for committees, so steering members can take a quick moment to either lgtm uh or add in any comments.

A

Are there any steering committee members on the line that do not lg that do not want to lgtm and would rather comment now uh or open discussion.

A

It looks like we've got a bunch of lgtm, so I have a feeling this will be approved.

B

Yeah, so I four by my account right now.

A

All right we're good to go on that one. Thank you very much committee.

B

Well, can we do an official? Can someone do an official lgtm approve, so we can merge that.

B

A

Thank you, y'all all right. uh Next topics we do have our an artifact hub issue. That's been sitting there for a minute. What else needs to be done here? Is this done? Bob? Is this sitting on you? What's going.

C

On with this it's sitting on me, I just need to actually log into the account and do it.

A

C

Take it as an ai for this week.

A

Is there anything that needs to be else that needs to be done with this issue before we close it all right and then the next one is a very big issue. That's been there. That's been sitting in community for a long time, which is the chair, responsive responsiveness, issue, we've done a number of different things um with this initiative, and I would like to ask steering because it does have a steering committee label on it.

A

When is this done? Is this done? What actions do we as steering need to take from this.

B

um So I would say a few things I think you know one we've got the annual reports which did not exist. I think at the time that this issue was created um which provides visibility around what sig leads are up to. um I think that you know over since we created this. We've also started to do the leeds meetings.

B

um It was as bob's mentioning in the chat. This was pre-steering liaisons as well. Liaisons have been shuffled and rebalanced across sigs working groups and user groups and now committees. um I think I think that you know to faithfully close this issue. We should just kind of scan through the comments to make sure that we kind of knocked out all the things that we said. um Otherwise this will stay open forever. So um I think if we, if we get someone to take points on kind of thumbing through each of these issues,.

A

Can I get a steering committee member assigned to this, so we can thumb through and potentially close and or open up new issues, bob wow, all right, all right, bob, all right bob, so you've got artifact hub and chair responsiveness, all right all right next section we have to assign some owners and assignments.

A

um One is to run down the uh cncf contributor travel funding, stuff figure out where we last left off yours on the call right now and then we also have lead training to get a quote start funding process, and then we also have.

A

The subprojects in sixxiaml issue.

A

Any any comments from steering or anyone that wants to work on any of those three uh three issues.

D

uh I'm interested in the sub projects um out of date, one that is connected to some of the annual report scanning sweeping stuff that I was looking at as well. Okay,.

A

That's awesome: oh oops, put you on the wrong one hold on put you on the right one. Here we go jordan.

A

Thank you. Anyone want to take the cncf contributor funding stuff or the lead training things I can help. I can help with the lead training how about that like? So, if anybody wants to pair with me, I can do that. I just can't run it by myself anymore.

E

I'd like to pair with you on it, one question I have is the um the timeline isn't clear. I don't know that I will get a whole lot done over the next three weeks, but I.

A

E

Definitely like, as the year starts, to be putting emphasis on.

A

I that's fair and I can do that as well, so tim in paris will pair and then um anybody up for the cncf uh contributor travel stuff and I gotta look at chat too.

B

I can hop in there.

A

A

And then, since eor's on the line um er, where did from from a cncf perspective, where do you think we left off with that? I mean at the last comment. Kind of has it dangling, but I I feel like there was other stuff going on in the background. So that's why I wanted to see if you had any more context.

F

uh I don't think that we really had so much stuff going on in the background uh because, like basically, we've stopped discussing it approximately the same time when we've stopped doing face-to-face events uh temporal stuff. Obviously uh so, uh right now, uh like the only major change we uh like, basically that we have right now is that, like we have relaunched our community fund for the scholarships, and currently it's named uh after den khan, and basically we have like this.

F

This brand new uh community fund uh initiative uh targeting our next, our our next face-to-face events, including the keep going in valencia, and so so. uh uh What I can do here like I can double check with folks from our events team and like folks, internally from from the cnc stuff like what is our current, take on this and um and keep keep you posted on this.

F

uh If you have any specific requirements so like, if you have like any like input or like anything, you'd like to discuss right now, I'm happy to do it or we can we can uh like. I can make some kind of like the background work on this one and uh we can discuss uh discussed it at the next steering meeting.

A

Yep well, steven is now your point person, so you and stephen can huddle and we'll figure out the rest from there perfect all right, so we are done with those two sections now sergey. I think you were. You were on the line earlier there. You are um go ahead and give us an update with dr shim stuff. Tell us how we can support y'all and what's good there.

G

um So yeah I already gave up there to view them a few of you, but uh just to give brief summary, uh we announced document removal, duration, duplication. Last year uh we plan to remove what this december uh before removal, we um run some survey to ask customers what's left uh and uh was they feel prepared and uh numbers were quite surprising, uh plus data doctor way showed uh that only seven percent of their customers migrated yet and like 96 percent of steel on docker shim.

G

So everything is points to the fact that, like something is uh needs to be addressed here and uh we need to uh work on that as a community. um Some comments from survey, like many comments, were about documentation unclear like give me best practice. What exactly I need to do step by step or give me a some recommendation for this specific project or for coupon mo like whatever projects they run on uh benefits.

G

uh I also wanted to point out that survey discovered that many customers had a third party dependency so, like we announced deprecation a year ago, and some veterans just adopted it and, like uh I know like uh dated up for windows, for instance, was just implemented, and uh some vendors are not like up to date here. So it's blocking some customers anyway. uh We discussed these numbers and results uh in um sig note uh just the last week before I think it was like.

G

Stick yes and um we all the signature believe that uh we still have time for end users, so we are giving another year for end user, while 123 is supported um one two, three one with the occasion.

G

We believe that third-party dependencies kind of chicken egg problem, if without firm date, nobody will migrate and without like we want to resolve it by by giving this from date and uh deadline. Extension unlikely will give us major benefits it will it may help some customers, but it may not help communities overall, because people will be less likely to migrate this extended deadline.

G

um So all that said, uh I also wanted to point out that we still as a community have a few things done. Not ideally like um uh we like vendors, adapting things very slowly. uh We have uh some documentation, tasks that we uh left out and uh like it wasn't open for a year, but nobody took it uh and uh we had like few attempts, but it didn't pan out. um We have some problem with how we call things like cri was just promoted to v1 and 123 before it was alpha.

G

So we basically saying use alpha think whatever like we know it's stable, but we call it alpha and end users. May I don't know how many end users know about it, but it may be hesitant to use alpha things uh and, lastly, like cuban, for instance, didn't even announce duplication and it still defaults to darker. So that's kind of um like we as a project like kubernetes, our cells didn't adapt and didn't migrate uh from docker stream and also promised uh backup story from mirages.

G

Now there is a court, but there is, as far as I know, no releases or no documentation. So it's uh also uh adds to anxiety from customers, uh but I want to remind the value of cncf one of the interpretation, with at least uh that we need to uh fastest better than slow. We need to move fast and help customers adopt better practices like container g cryo is a better c rise and docker integration.

G

So we believe that it needs to happen like signaled. So in order to do that, we we wanted to help customers at least to make some attempt to uh streamline their migration efforts and help with unknown issues and uh problems they may have.

G

So we already had a meeting with celeste and bob was there and like few other people um about comms and what tasks we need to complete. uh If you want to help customers what we can and cannot do and uh out of all this, I wanted to uh formulate a few asks for steering. So first, I think uh out of all the tasks we identified. Many tasks are documentation, tasks that potentially cncf can help.

G

This uh practice shows that uh out of signal contributors, for instance, or like uh random contributors, we didn't get uh proper results. We tried and like we didn't, get all the documentation we needed, so it would be better if cncf can sponsor uh this writing this documentation and uh potentially like helping end users with like case studies or like educational series, or something like that. uh Okay, thank you also.

G

uh I I think we want to uh publish some empathy block to end users. They spoke up. They said that they're not feel prepared, so we need to reply to that somehow, and this message I believe, needs to go from uh steering committee, so steering commission used to tell we still doing it, because signaling architecture thinks it's a right thing to do and we will help you to uh through, like scenicity of engagement, and I will have to do others. Documentation uh updates.

G

uh Also. I want to make sure that we understand that, uh like 123 will be very popular uh version for a long time, especially if you forgot something and some uh customers will be stuck with docker shim. So potentially we want to ease end of line for 123 and maybe even extend it proactively and lastly, I need to. I think we need to double down on this issue on community repository, asking about formulating how we announce big things, and I propose some ideas.

G

What we can include in this uh announcement, but I mean it's a really big issue and uh needs to be unlocked. So I'm done.

B

Yeah, so my notes from you know part with you know, wearing wearing the steering and wearing the sig release. Chair hat, um we also had a request from the sig release side that sig release send notification and assist with messaging. um I strongly feel that you know our arch and node are in the driver's seat here, um and the notifications and messaging should primarily come from them. um With regards to the release cycle, we are not extending a release cycle.

B

The the release cadence has been modified, already um kind of across the board um to give buffer time for exactly these kinds of situations so like at least from the sig release perspective. I I don't plan to. um I don't think we'll plan to extend a release or end to life.

C

uh I guess the as far as the the commitments goes from the we we do have commitments from the cncf already.

C

um I don't know if it's been act by um chris, hey on the tracking issue back and dig that up, um but yeah yes, they are celeste is helping, and then um I forget, I'm told by god his name. I'm sorry, the the other uh big technology.

H

From the scene.

C

Yes, thank you. Thank you. uh Nate will be helping as well um to help coordinate all the the updating of the documentation.

B

So so an additional upside nate is actually on the release team and I believe he plans to also be on the release team for 124, um so that'll be a pretty tight interlock there um josh just to address your um your note in chat um talking about the extending the release cycle, uh not the release length, but the patch period um we've already extended the patch period to allow for maintenance mode.

A

And then kristoff you're the last one on this topic and then let's wrap it up.

I

Yeah, I just wanted to to uh make clear a point like um the sergey, like uh I, I think I think, from the steering side uh we support uh sig release and sig arch and the other groups here for making signo for making the the the decisions that you are as far as like deprecating this, but I think from a communication standpoint, just for clarity's sake, like uh steering committee, doesn't typically take on uh technical decisions. uh They're like very specific technical decisions, like all those are delegated out to community groups, to make those decisions.

I

So we support you, but also I don't think we're going to take lead in any of these communication pieces and say like steering, is saying we're doing this, um because I I think that's not the the the pattern that we typically follow in the community. So we do support you, but the way that we're supporting you is backing you up in the in the forums that, like signaled and saying, architecture and sick release can like make make a decision like this on their own.

B

uh So so I'll, sorry, paris, can I jump in real hot, um so yeah we also the release leads, have a sync on mondays: we discuss the docker shim stuff as well. um I think what we can do in terms of messaging or what we should do in terms of messaging. It's a very uh you know, proactive stance in terms of like making sure the deadlines are hit and the with the stick being that the release will not go out. Unless these docs are right. The release notes are right.

B

The like people are prepped for it like this is. I think this is one of the situations that it does make sense for sig release to block a release if it if it is not up to snuff.

D

So, to be clear, that would be the 124 release when it's actually removed. Is that right? Am I understanding it correctly.

B

A

All right- and I'm also going to take an ai to work with the end user community for cncf and make sure that this is on their docs and work with bill. So all right, sergey sorry to russia. We have a tight agenda.

G

Thank you very much.

A

Thank you very much, um josh you're next, with election stuff.

H

Okay, a bunch of election stuff, mostly issues that um shouldn't be too hard for steering to decide um the.

H

So we've already had the election retro. If you missed it, you can see the notes. um The um first question is an issue that came out of the last election retro last year's election retro.

H

That was never resolved one way or the other, and I just want to make sure that steering wants to keep the rules the way they are. This concerns how we handle things when we actually have to implement.

H

Representation limits.

H

Currently, the way the process works is, if you have to implement representation limits because of a group of three people, one of whom was elected the previous year.

H

um You filter one of the newly elected people um based on preference voting right.

H

um One of the things that we inquired last year is whether or not we wanted to give a chance for all of the members of that organization to meet and decide themselves um who would drop out, particularly in the case of, for example, an incumbent steering member who was having trouble keeping up um the um um uh the. I just want to close this issue, one way or the other um if we think the way that we do it now is fine um and we don't want to change it.

H

um Then, uh let's just have somebody from steering say someone close the issue, um and if we do want to revise it, then um I can come up with some revision language.

B

um I can take point on this since it's near and dear to my heart: okay,.

H

The um so yeah either either way, because just right now it's just open and it keeps getting extended because nobody said anything one way or the other. um The other two issues would be rather more straightforward.

H

The first one is.

H

The steering committee candidate uh criteria- currently there is no requirement that somebody be a um that somebody be um qualified to vote in order to run for steering um and within the last two elections that has run into a situation where we had a candidate um who was not able to vote for themselves.

H

um The um that was weird it didn't necessarily end up being you know some kind. It was not like a technology or process problem for the election, but it did seem very weird and uh wanted to raise with steering whether or not we should continue having that criteria or whether we should require that a candidate be qualified to vote, whether via contributions committee membership or exception.

B

I personally feel like a candidate should be uh eligible to vote and and run, and that the you know the the voting eligibility should be some subset of the the candidate eligibility.

A

I

So the the the way things are right now is intentional, uh at least as far as for historical reasons, that doesn't mean we can't change it, but uh just to kind of bring some of that historical context.

I

um When the bootstrap committee was was discussing um those like initial criteria, parameters for like uh who can run and who can vote, um the thought was you can restrict, uh who votes you can restrict? Who can run? You can restrict both, but both may be unnecessary um and, in particular the thought process behind the current requirements are um there may be, um there may be candidates that are?

I

Are you know not necessarily directly inside the community that may uh that there may be a reason that the community wants to bring in somebody from the outside, whether that's they they want outside perspectives or or such, um so they, the restriction that that was initially placed is restricting.

I

Who can vote um was like a clearer line to draw, um but then leave the the candidacy piece uh open um and that anybody can end up running if they can show that they have nominal support through the the nominations process, but uh yeah anybody can run but then kind of restricting it on the voting side uh to determine that, like the community itself is determining who it's, who its leaders are where those leaders are from the inside or the outside.

I

um So with that context, I'll say my personal thought is I like that original context, uh and my current thoughts are. We should leave things as is even if it might be. It might seem weird on the surface that I think there's like good reasonings behind why it is the way it is.

A

Bob and then we're gonna move on to the next uh election retro item and then steering. Please weigh in on the issue that josh has set after.

C

My personal gut take is that the person the people that are running should at least be an org member.

C

This essentially requires them to at least be on our mailing list and keep up to date with the community, um which you know, isn't necessarily the same as our current voting requirements, but at least ensures like a a a very low baseline, to be informed with the community, um and if we wind up not going with this option or, like you know, further restrictions in terms of um being a potential voter to be able to run as a steering committee member, we should at least take the the historical context and pr that in some place, um because the question has come up now, multiple times.

A

All right so steering says is the first time that we've seen this issue. Please vote on the issue and if it's not voted on by the time we get to our next meeting, we'll add it to our meeting log under vote and we'll vote there cool all right. What's the next one, josh.

H

Okay, uh the other one is a little bit more straightforward in a little bit more obvious in terms of people being four at all, it looks like we need to actually refine the criteria, which is um one of the things that we've noticed over the last several elections is there are certain groups within the community that are active participants in the community in a way that doesn't show up on github, um one of the paramount examples being the code of conduct committee and we end up processing a lot of exceptions for those folks.

H

So the original proposal from the election officers was that we automatically include the members of the code of conduct committee and the security committee as well in um the list of eligible voters, whether or not they meet other thresholds.

H

um I, after that, um it got suggested to kind of expand that to anyone who was mentioned in cigs.yaml, the.

H

Any of which would be fine, which is basically my worry, obviously, is putting somebody in the position of chronically of asking for an exception. Every year can result in them failing to do so or failing to do so by the deadline and then not being able to vote. um So it would be if, if we know they're going to ask for it every year and it's going to be granted, we might as well put them on the list of voters.

A

Stephen, we have, we have room for one steering, comment and stephen, you have your hand up and then we'll do the same thing since then. This is a new issue for us, we'll either vote on the issue and if we don't have enough votes by the next meeting, we'll vote at the next meeting, stephen cool.

B

Yeah, so I yeah, I do feel that at a baseline, the committee members should be included, um I think, for I would be in favor of sigs.yaml if we could guarantee that zig.yaml was up to date, which means kind of going through user groups and and any other groups that may be potentially out of date. As a result of you know, annual reports or some sort of metric that we can find.

H

B

H

More discussion on the issue, then uh final fyi, to close out my item uh within sig contributor experience, we're actually forming an election sub-project because we need to have some people who are looking at elections, issues between steering committee elections as well as enabling sigs to run their own internal elections.

H

So you will see that coming as a subproject proposal later on.

A

Awesome, thank you so much josh and thank you so much to the entire uh election officer committee that we had this go for for all your work. Thank you all right. Next up is our favorite topic annual reports and then we've got sustainability plan. I'm clocking like 28 minutes. If that, uh on the rest of our meeting so annual reports, we've got a few things. We've got wrap up from last time uh and then we've got dates and deadlines and we've got who's doing what so.

A

um First it looks like most of the wrap up is listed here in the agenda. As far as our comments, I don't think we need to necessarily go through each one, but does any steering members have any comments or questions about the wrap-up that they would like to discuss here.

A

So it looks like kristoff can put the time in for the generator within the next month. However, we've got some contrib x mentees that might take that, might take it further quicker, um jordan and steven. Do you have any updates yet on the questions or not yet.

D

um I left a comment on there. I had gone through the the questions we asked and pulled out about 10 that seemed like we have data for, or could dry or pull answers for automatically um I sort of categorized them as like. We should be able to answer this question completely automatically or we could at least pull out it's the yeah very last comment.

D

We should we could at least pull out like a partial answer, like a skeleton to say like instead of saying what are all the kept things you did this year be able to say, like we scanned the repo and here's the kept things you did this year, please like add anything or add color or explanations, or things like that, instead of making people go and everybody construct these things manually.

A

B

Yeah and- and I think I I think jordan gave a a good assessment of um on this issue comment. um There are some particulars, I guess, for you know, meeting recordings up to date like how do we determine that um we can pull the playlist? We can easily pull the playlist but like if the playlist is not up to date like how are we getting that data?

B

So I think I think you know we can dive into some of these, but I think I I really like the way that he laid a bunch of this out.

A

And uh jordan, would you have? Could you take point on peering this into the policy like with? Maybe any potential changes to the questions um that you have here outlined.

D

Yeah I um yes, I can change the questions. Obviously, for anything like this there's there's then work to say like here are the questions we think can be generated automatically.

D

Here's the ones that only a human can answer do we want to just divide the list to begin with and say like, let's get humans answering questions, we know only humans can answer and in parallel, try to work on like spinning out individual efforts to gather these things and fold them in because if we wait until all of this automation is done, yeah like we'll never get started like actually getting human, yes, yeah yeah, yeah yeah, and if you I'll sort of just divide the list in half and say like this is stuff we're working on automating.

D

Please answer these like five things or whatever. Okay.

A

Yes, yeah. That would be awesome any oh tim. I see your hand and stephen. I see your hand as well tim. First, though, I'm.

E

Wondering if this could be initially just a simple thing on the liaisons checklist, something that you're watching for periodically. I I question that how much will be automatable? Because, yes, you can? You can validate the the positive case like in the last three weeks. There's been something uploaded, but if there hasn't been something uploaded, maybe you have to go back to the the median calendar and discover that there was a conflict and that meeting was postponed or there's in the negative cases.

E

There may be valid reasons and again that it's going to be on the human. So maybe a lightweight checklisty thing that the liaisons are supposed to just do regularly.

A

That would be a good issue. Could you file that time.

A

Thank you and steven.

B

Yeah I I kind of agree there too. I mean, if you have a recent meeting. Upload, doesn't necessarily mean it's a recent meeting. Maybe the meeting was just uploaded, but it's a meeting from many months ago or something right. um So stillness is important, but I mean also the uh what if we were to ask people what they're working on in terms of of discovering some of these things right.

B

So I think you know there's one around looking for looking for stale uh sale owners and you know, dims has been working on that maintainers tool, uh that kind of scans, a repo for owners and pops people out, compares it to dev stats pops people out. Maybe there are other tools elsewhere that we don't know about that people are working on. So what if this was a question as opposed to just uh we're working on automating? It maybe you've already done it.

A

And speaking of bob actually has something to the next uh effect on the um on the agenda. Bob you're talking about a tool to crawl, shared repos.

J

C

I don't know if there's an open issue for it. It came up in like the the mentee uh slack and some other conversations regarding scanning the shared repos uh to basically provide an updated list based off the like sib labels, sig sig labels in the owner's files that could be used to update like uh six dot emo, because a lot of those are out of date um and that would also feed into dev stats. Because right now, devstats is inaccurate for shared repos and when it comes to tracking the sig specific work.

C

There's a long-standing issue with them. They cannot view prs and things like that by label. It has to be by like file path for them to associate um potential prs and commits with a sig.

J

All right, I've got an action item for you to set that issue. Okay,.

A

All right- and uh I also need we also need to wrap up whether or not we care about working groups reporting into their sponsoring sigs, because the last um the last annual reports netted us the the knowledge that most working groups are not giving updates to their sponsoring cigs um and how we want to deal with this. So I think one one. I think the action should be to take this into a new issue um and discuss it there, and then we can discuss it at a later meeting. But stephen go ahead.

A

You have a hair enough.

B

uh Yeah, I strongly feel that they should otherwise. The sponsorship feels hollow, um and it also kind of kind of you know, spreads the work out.

A

Okay, all right, so I'm going to take an action to take that into a separate issue for us to discuss there and figure out what to do with that, and I know tim you come. You had a big comment on that as well, um so I'll make sure to include your comment on the issue as well when we set it all right, so that takes us to I'm sorry, stephen. I.

D

I'm just going to say just as a like sort of worst case scenario. Like I mean if working groups are doing an annual report um tagging their sponsoring sigs into their annual report, uh that wasn't really a thing we did last year, but thinking about it I mean I think a year is longer than I would prefer, but if it's better than like, never um at least saying like every year like here's, our look back and I look forward to what's going on also hey sigs that we supposedly like file up to like hey.

D

This is what we're doing did you know? Do you have feedback could be a good starting point.

A

Yeah, oh man, my typing skills today are just whack hold on I'm having an issue with notes right now, y'all all right and then stephen. You still you had your hand up, I'm not sure.

B

It was up for the last one: okay,.

A

Okay, I just want to make sure all right, so that brings us to dates and deadlines for our next go, which let me share my screen.

A

So we need to decide pretty much on a couple. Major deadlines, so that'll push us to push us to get the rest of this stuff done so meaning, like I don't know. If we want to think about working backwards. um Last time we produced it. I think it was in june, which isn't really much of like a year-end wrap-up.

A

I think we should uh publish it no later than march 31st. um Does anybody have any comments or just want to start discussion around uh when we should publish and or start this process.

A

B

Hello, hello again, um so I you know, I'm I'm just thinking about 124 a little bit and how that all already seems like it's going to be a contentious really like. Are we sure that we want that to be the reports to be published around the time that people are kind of crunching.

A

And should we bump it up to january.

B

B

Of way of what? Year of.

A

2022, like that, like january 31st, I feel like, because I think, in this project we're always going to have something that we bump up against whether it's kubecon, whether it's a release like had bob. You had your hand up.

C

uh I was mostly plus one for some time in march. This also like one of the big problems we faced with last year at the annual reports is because it extended so long people kept mixing up the current year with the past year. uh When it came to things like, we need people to really start thinking about.

C

You know what they did in 2021, not 2020, half of 2021 plus half of 2022..

C

It might be actually easier to phrase it, as what did you do in these three releases?.

A

Stephen, did you still have your hand up from the last one, and I see jordan? Well now, jordan beat you to it so jordan. First.

D

I was just gonna say um I think, because pretty much everyone has done one of these before, uh like start like looking at what they did last year and then looking at deltas, it's going to go quicker, especially if we can take some of the questions and like move them to a.

D

We should have a dashboard for this focus on. Like these questions, I think getting human answers to like the deltas over last year's report should be pretty easy to do like in a month. So targeting january is bananas because, like half of the people are either actually gone or like mentally gone for the rest of the year. At this point, um but targeting end of march seems very doable to me. uh I- and I would I would when we asked people to do this. I would encourage them to say, like look at last year's report.

D

Think about like what changed and like what's gotten better. What's gotten worse, highlight those things um I think it'll be less of a lift than it was last year.

A

Other comments from steering.

A

Don't care just get it out comments.

A

Tim, my fee, I feel like you're, you you're you're, on you're, on the cusp of a comment right now. I can see it like.

E

It's definitely like I mean annuals in the name right, it's good, you gotta have them happen and and deferring doesn't help, but also you want quality. So I don't know.

A

All right, so how about this? How about can we take an ai that we need to have all questions policy procedure pr'd in by january 15th and then um then january 15th liaisons start their process with hey y'all? Remember this annual report right and then, like you know, we'll give. Maybe we can prepare the liaisons with like a checklist of like what to go over blah, blah blah and then from january 15 until mid-february, that's like chairs getting prepared and then um and then like.

A

We expect it either like we'll, I mean we can vote on this at another time at another time, but maybe like end of february or end of march, it sounds like what we're gonna vote on for due date.

A

And then so publication date, we should try to aim for no later than mid april.

A

All right steven han.

B

Oh yeah, I was just going to say: let's, um let's chop it up, maybe on the issue. Everything else.

D

J

D

J

D

Right um concretely, for that first step like when we have our stuff done, I I like january 15th or even earlier, like I think, end of year is yeah um and that's the one like we're. We may or may not have a chance to meet again like this month, so yep the the thing that meets in the next year. We can kind of finagle and discuss, but if we have a crisp date for us, that would be good.

A

Yep exactly so let for right now, let's make it june 15, but let's try to target and the year and then we can probably move faster, all right so stephen hand up or down right now.

B

I'm so sorry it's a guy.

A

All right so this this part, I won't lie to you all the who's doing what I did not do a good enough job last time, because I did much of this myself. So I didn't do a good enough job of like cutting up roles for this, but I definitely think we need roles. I ended up. This took me about 80 hours last year, so 80 hours, meaning from prep and outreach and the actual summary um it's too much for one person is what I'm getting at here.

A

So I think we somehow need to cut this down into like who does what? Obviously, we've got liaisons that roll this out to the chairs, um but I'm thinking, maybe we have a point person. That's doing point for like policy procedure prs like making sure that everything's good there, but I think the biggest part was the summary um that took at least 40 hours, um and that was really painful. So I don't know how. Obviously we don't have to talk about that now.

A

I just wanted to put that on the agenda as something that we need to think about right now um and like how we carry this out and how it's not uh you know a one-person job. Maybe it's a three-person job whatever it is, um but right now we don't. I don't have a good enough, like you know, role summary. If you will of each of these or, like you know, know how we should carve out roles, so I kind of just wanted to discuss with you all right now.

A

A do you think we could trim out rolls from this b. Does that make sense and c just how do we cut this work up?

A

D

Bob's comment uh I feel like there was a lot of sort of interpretation and synthesis required last year and if we can rephrase questions so that they copy paste more naturally into the summaries that might save us. Oh yeah excuse, oh wow. This is recording yes,.

J

A

um And then bob.

B

uh Actually, I think, stephen, your head your hand uh yeah. uh I actually have my hand, so I I agree with the copy and paste uh technique. um I think that you know the the liaisons can handle roll up for their their sections. I think that makes a lot of sense. um It makes sense to chop up the you know: prep outreach, I think you know having people on point for uh one uh prep and outreach and then multiple for summary.

A

All right and we can work on summary stuff once we've got all of our stuff pr'd in and you know our policies, procedures and questions and stuff like that. Bob.

C

So one thing that will make the copy pasting easier is, like everyone decided to format their uh reports like in a different way. So, like the more we can like sort of provide them a template of how to answer these questions. Even just like you know, making sure that there's bullet points there will make that process easier.

C

So it might be in addition to this just review the template itself for copy pasta, ability.

A

And also the generator as well.

C

A

All right- and I think we're done there any other questions- comments, concerns uh discussion, points for annual reports.

A

All right, I tried to get through that as quick as possible. Sorry y'all um next is sustainability plan, and this is a nice lengthy document that is um making its rounds through uh governing board folks right now, and um what I'm trying to do here with this document is determine what the needs are for our project as far as like what we need go forward as well as what we need now in order to be both uh stable as well as sustainable, there's a lot under needs.

A

However, I don't think that this is all of our needs, so I would love if liaisons can work with their sigs on what sustainability means for them and also what it means today, as well as what it means in five years, um we have a ton of needs right now.

A

uh However, the lot a ton of the ideas that we have for how to address these needs doesn't even necessarily hit all of those needs. So um what I wanted to hear from steering as far as discussion is how we can get all of the information that we need from sigs in this regard and if cigs don't tell us like, for instance, if cigs say that they're fine does that mean that these people have a plan or should we just keep and should we keep it moving um like?

A

How much should we get involved here when um six things sig say that like everything's good for them,.

E

I think that ties into the issue, I'm opening around um documenting the layers on rule if we have some of those criteria defined there in a document, maybe it's more clear.

A

Stephen and then bob.

B

Yeah, I think, I think, uh to determine whether or not cigs are good. I think it's less, you know personal opinion from them and more so. uh What is our baseline of good right? Is it? Is it meeting the standards of the the annual report? Is it you know something that steering has seen or something uh outside of the community or within the community that that people have seen like what I'm not sure that we have the baseline um to determine what good is.

A

C

So I thought it was sort of along the same lines in terms of a baseline of what is good but like it's, it's also sort of I'd say, got feeling from the liaison and keeping an idea of what you know how how the the group is. You know functionally in the community.

C

It might not just be from the liaison, but just sort of like a gut check and how they are interacting with everyone else, and that is sort of play as a responsibility to keep the you know interact with them um and as far as potentially surfacing some of this stuff more, um maybe we make it a part of the annual report in terms of what you know, what jobs, what what things people want to see like where they like.

C

I know we have where they need help, but maybe try and get people to flesh out more. So it could be more integrated into this style report or style thing that can be fed up. You know each year.

A

And then do any steering committee members have ideas for how to get some of these needs covered.

A

So, for instance, um I'm not gonna. We don't have time when we have literally four minutes, so I can't go through all of these needs right now, but we've got kate's infra, sig testing, contributor experience, docs cloud provider, release instrumentation, node and then also some critical staffing or some critical dependencies that need to be staffed.

A

um There's a lot on here and there's a lot that I think um needs attention. um Stephen.

B

Yeah, so I think you know part of it is the the contributor to kind of maintainer path, a lot of the time that is being spent by maintainers. Today, I think, is being spent on. um I think it's maybe wrong to call it minutia right, but there there's a bit of moving things around.

B

So I I like seeing the program manager and the and the community uh kind of architect roles uh listed here, um but there's also, you know if I think of just looking at the sig release stuff, for example like getting full-time release, managers right. Those are, you know, people people who have spent multiple cycles and you know getting getting up to speed with our content and and our tooling. um So it's like it's hard to.

B

You know from the maintainer aspect, it's hard to just like stick someone in there. That is immediately going to get it right so, like how do I? How do I take a contributor? How do I take an existing contributor, bring them up and find a role for someone who's coming in new? Is it would be my concern as a sig chair right.

A

What if we had a cncf kubernetes fellow and they were there, their full-time responsibility was releases as well as test infrastructure.

B

Yeah I mean I, I think, that's potentially workable. It's still it's still a matter of like what is the level of ramp up time. That is going to be able to take them to be effective.

A

And then um yeah, I'm going to put in a funding request for k-tempra and sig testing.

A

They are at a point where new contributors are not getting on boarded, so we need to get them some better documentation and some on-boarding programs and some like hand-holding programs with contributors to help them um any other steering committee members that have discussion topics on some of these needs.

A

Okay, well, that's it for oh stephen.

B

um I think it would be. uh You know, I think something you know analyzing the some of the boards that we have today with dev stats right, the the time to time to merge time to review things like that.

B

um Because again, I I do think it's enabling new contributors to hold these like full-time roles, is going to be largely dependent on freeing up existing contributors to train them up like if, if it's, if it's a matter of like oh, these issues have not been seen and or haven't, been touched in months, because no one's doing triage and because there's an expectation that the chair is doing triage.

B

So the tech lead is doing triage but something's on fire and that's what they're focusing on instead, um it's hard to like the triage is it's it's not easy, but it's it's a good on board right. um The okay. If something is, is completely broken, you have to jump in and it's like. 2 a.m is a harder thing to do. Right.

A

I think most of the sigs that I that have needs are have needs for projects that are like 40 plus hours like refactoring reworking. Something like I mean, and I think in kate's infra, it's we're officially over budget, and how can we knock it over budget again? Well, we can probably do that if we, if we rework prowl um and cut some of the cruft, I mean that's just like that's just an example.

A

Right, like I feel like all of the needs that have come through are needs that are not getting addressed quickly by new contributors. um So that's the. I think, that's one of the biggest themes that I see with the needs that have come through.

A

Most of them are asking for, like full-time type of full-time type of individuals to be assigned to them, so um so yeah right now, the ass to the governing board has been. Please send us your full-time folks uh and I've been doing one-off meetings with individuals like oracle uh and many others to try to get more folks onboarded, um but as we all board folks they're, you know having an issue getting you know getting acclimated into into some of the cigs. So.

A

All right, it's 10 30.! We got to wrap this um there's so much more to do on this, but liaisons. Please work with your sigs on making sure that if they have needs that they're addressed and that um we start getting, some solutions addressed or solutions assigned to them.

B

Thanks everybody.

A

All right who's, the after the the private meeting, who wants to be the next uh bason for jeff for january.

A

D

I'll take both in january.

A

Okay, all right, although.

D

You're an excellent example: paris, I'm running a tight-knit.

A

I didn't I didn't I didn't because, like we didn't get to like you know the 800 other things that were on the list, so all right see, y'all have a good week.