 Hey Ted. Hey, how's it going man? Finally going okay. How are you? Doing good. Yeah, I heard you were sick. Sorry. And you know like normally food poisoning lasts for a day or something and I was out from last Wednesday afternoon, like late afternoon until this morning. It was terrible. And then I got a near infection at the same time, which was just like the cherry on the top of the pile, so yeah anyway, doing a lot better now, which is nice. Good to see you. Let's give people a few more minutes before they join. Well, I mean you're back. Yeah. I just got back this morning. I came back last night, so Morgan where are you coming from? Me? Well basically the toilet because I had food poisoning. Sorry to hear that. You know that this is on record now. Anyone who emailed me in the last week would have gotten the auto response explaining that I was sick and whacked. Let's start at 4 o'clock, that's the hour, like someone nicely already got the agenda going. M says they can't hear, I think it's just them. No, yeah, because yeah, it sounds okay for the rest of us because we're all talking to each other. Oh, they fixed it. First topic, is that going to be Constance, I'm guessing, because I know she sent an email about it. No, that'll be me, I actually think Constance can't make it to this meeting or at least not the first part. Yeah, she can't make it to this meeting. Oh okay, so that'll be Ted. Perfect. Tell the clock strikes, strikes four after. It's 4 o'clock, so welcome everybody to the monthly open telemetry community meeting. This is where we get together, discuss status and go over any other community news or community issues. A lot of us just spent some time together at KubeCon, which was pretty exciting, so there will be less to discuss at this meeting than usual, but that's okay. We have the agenda shared in the calendar invite, so if you use that to join, there's a link to the agenda there. And the first item on the agenda is Ted, community feedback. Ted, you want to pick that up? Yeah, absolutely. So you know, we're continuously evolving and growing this project. We evolved our RFC specification process over the summer, but we haven't really had like a big community check-in about the general sort of project structure. So you know, the way we're kind of set up is there's a governance committee that creates a charter for a technical committee, which helps manage a bunch of SIGs, special interest groups, and each special interest group has a set of membership tiers and contribution requirements needed to reach each tier. And so it would be great to just get some feedback from the community about how that process has been going and how we could improve it. And it could really just be general purpose feedback as well, if there's just something about the project where you're like, oh, I see this thing and I wish it was better. And you prefer to submit that feedback through an anonymous form. We're providing a form for that feedback as well. So the people who will be reading this feedback will be the governance committee. You can optionally include your name and email if you'd like to have a follow-up conversation, but it's also fine to include anonymous feedback. So it can really be about anything, but we're primarily interested in the structure of the project and ways our structure or process could be approved. Any questions about that? Ted, I would like to make more advertising of this. Maybe post it on Gitter, and yeah, Gitter is probably the best thing, and everyone who is from other SIGs, please make advertiser of this form in your SIG meetings so that everyone has a chance to know about the existence of this and provide feedback. I think that's great. We could also post on Twitter, but it's probably a terrible idea. No, no, no, let's not do that. Unless you post it on your Twitter so that we get all the feedback for you. No. Other comments on it? Next topic is from Chris, target release date for clients. Yeah, so right now we've got December 31st as the release date for the B4 for all the clients, and that was always kind of a placeholder. I don't think we actually expect to release anything on New Year's Eve. So I'm just curious what everybody thinks we ought to do here. Well, I definitely think people, we had in earlier days, discussed trying to keep the version numbers in sync, and then I think we eventually decided that that wasn't realistic to do. So I would say individual SIGs, at least personally what I would suggest is individual SIGs just continue to ship at a regular cadence. In terms of versioning the spec, I'm trying to get a version of the spec hopefully by end of next week that seems tight, but I really would like us to be shipping a .3 release of the spec and to have that contained. But I think are sort of the last issues, modulo, the kind of named Tracer resource stuff. But that seems like that'll land well. But obviously, if we're releasing that spec at the end of next week, I wouldn't expect people to be implementing that over Christmas break. Why not? I already have my New Year's Eve party planned. I'm sitting and working on open telemetry. Yeah, can't imagine why not. So your suggestion is that each language, say, come up with their own dates, likely, I'm guessing, in January for most of them. Yeah, I mean, but I would say keep shipping versions. Don't worry about the version number as far as improving the spec. I would like this .3 release to ideally be the last major changes. And I don't even know that the changes actually are getting narrowed down. So I don't think it'll be too invasive when we do it. But my hope is, in general, in Q1, starting in January, we'll be able to start pivoting more towards productionizing what we have and putting it through spaces, writing instrumentation for it, and stuff of that nature. Yes, this is my second question about the specs then. We don't right now have a schedule for beta and GA. And I know we've talked about having a GA release in March. But one thing I think we haven't really described is the tests that we want in place before we call our releases GA. And I think this is what we need to focus on in January. So as we have the final version of the spec and we know about what every client needs to implement, I think it'd be really helpful to also have the set of tasks and guarantees that we want to call the releases GA. Yeah. Maybe like the beginning of January, we can have some kind of like kickoff meeting to specifically brainstorm around what is the checklist that we want to go through to make sure, like, what are the specific things that we're looking for to say, like, this is a GA release? Like, what do we want to test? What tests do we want to have in place? What's a good form for that? Is that something the technical committee should do? Yeah, I think it's fine to be part of the technical community charter to define that. Yeah. But maybe it's like getting that plan together and then presenting it to the community and getting some community feedback. It would be good to have a... Maybe there's a community meeting that's timed in like early January where you can... Yeah, one hours is January 8th at 10 a.m. PST. We just want to do it there or is that too early? Or January 8th may be tough. I would probably... But do we need to do that in a community meeting? No, not necessarily. It could be that or the spec meeting, but it seems like maybe the TC can get a draft together early so we can get around to feedback in early January. So... OK, so let's say... Let's plan for this. Let's have the TC get something together and then we'll get early feedback by posting it on Gitter in the mailing list. Just like get written feedback. Does that sound good? Sounds great. OK. How do we get this like onto the... I'm not... I don't go to the TC meetings. How do we make sure the TC actually takes action on us? Bogdan, do you want to... Yeah, I will bring it. Thank you. All right. Any other thoughts on this topic? Moving on. KubeCon summary for the community. Bogdan, you added this. Do you want to speak to it? No, I was curious if we should do this in this meeting and if, yes, probably you are the best person who wrote the... I mean, I saw the write-up. It's very good. But let's rephrase the question. If anyone has any question about what happened at KubeCon for people that were not there and they don't know what we did... Or didn't read Sergey's excellent write-up. It was a great write-up. It kind of covered everything. Is anyone on the call have any questions about KubeCon or want to go over the different things that were presented? I guess where is the write-up shared? I would like to get access to that. Yeah, I actually pasted the link into the notes on this talk. It's also just on the OpenSlimetry Media blog. It's the most recent post. So, yeah, Sergey summarized the keynotes, the talks that we had as part of the community and also the community meet-up that we held. And finally, the Observability Summit, which Ted had organized. Ted and others, I think, had organized. And, yeah, it's a very thorough write-up. Speaking for myself, having done two of the talks there, I think it was a big success. And, yeah, I was really excited just to see how large the community has grown and how often OpenSlimetry is being referenced in talks and other projects and things that we didn't even set up. It was super, super awesome. Yeah. I saw a write-up in Japanese blow-by just looking through Twitter. So we're big in Japan. That's always a good sign. Yeah, it was amazing, especially the Observability Summit, seeing how people were already using OpenSlimetry or other things that I think a lot of us have worked on, like trace context, in ways that we hadn't originally imagined or ways that we hadn't really heard about until we went into the talks. So super successful. Very excited for the next KubeCon in Amsterdam because we will, with any luck, be going beta then. So it should be a big surprise. Yeah. Yeah, who knows? So, yeah, looking forward to that. Also, we're gonna put a plug-in and say the talks are all now online, too. So if anybody's interested in watching them on this call, you can go find them on YouTube. Oh, yeah, there's YouTube videos. Yeah, and they're accessible on the CNCF website. And I want to say that Sergey also may have linked to them in his post. So we'll check. Yeah, they're all linked in the post. Yeah, perfect. So yeah, definitely check that out if you weren't able to attend KubeCon. It's an awesome summary. And speaking of Sergey, the next topic is yours, Community Bridge Projects. Yeah, I just wanted to highlight your attention. If you want to be a mentor of some students, I'm not sure all the details about this program of Outdoors, exactly, but my understanding is CNCF will sponsor five projects. Connecting mentors from CNCF projects to students who wants to get some experience. So you can either take one of those projects and put your name as a mentor. I'm totally fine with any additions or put in more projects there. These are proposals for projects? Or these already have funding? Proposals. This is a bunch of proposals. And CNCF somehow will pick five. I'm not sure about the process yet. Oh, wow, people already, Sergey, I'm guessing you've pasted the ones that are already on the page. Because there's already several open telemetry projects there. Yeah, I put them there. Awesome, the CNC pages would be great. Agrations registry, nice. It's all the projects that are not on a critical path, but it's very important for our community. Yep. Morgan, in that same note, should we try to propose some of these to GSorg? Yeah, then we have in previous years to Google Summer Code. Yeah, we should definitely do that. I suggest maybe we just use the same list. Yeah, using the same list. Maybe the ones that the CNCF does not approve and whatever. Yeah, that sounds good. When does that happen? I know we do it every year. I just forget what time of year it is. I don't know exactly, but it's when you- They'll email us. Yeah, I'll take action when I get the email. Okay. There's a great list. Thanks for coming up with that, Sergei. And yeah, as Sergei said, if anyone has additional ideas, please add them or if anyone wants to mentor. I can think of a few people for Z-pages, for example, from Google. Then please send yourselves up and mentor. All right, next is SIG updates, starting with the collector. I think I saw it. Tigran, do you want to talk to Pet? Yeah. Yeah, give an update. So our focus recently was primarily on two areas. Number one is improving stability and making sure the collector is ready for production use. Number two is reviewing and accepting new components from other contributors. AWS exporter, Azure exporter. On the stability track, we've recently added metric testing to the collector testbed, in addition to trace testing that already exists. And we plan to add more end-to-end tests for protocols supported by collector, add more of those automated tests to the continuous integration builds. This is basically part of the effort from Splunk that we announced recently that we want to contribute more to the collector. And we plan to use the collector more widely in our observability products, instead of proprietary or other open source agents we have. So we want to make sure it's stable and good for production use. And the other part of this effort is contributing functionality that we have in other agents, which we believe is useful for open telemetry community. Another important update on the collector is regarding the stability of exported interfaces in our code base. So we were planning to announce that public interfaces consumable by third-party dependencies are stable. However, this announcement needs to wait until we declare the open telemetry protocol as final and improve either specifications seek. Since we plan to make use of the protocol schema in the internals of the code base. So I just want to make sure that all of the contributors will develop components for the collector. We are aware that these interfaces they depend on they are not declared stable yet and may change which may require changes. Well, we'll do our best to minimize interface changes, but so contributors need to be aware that breaking changes are still possible. When do we expect the protocol to be final? Whenever I think the specifications seek decides that there will no longer be major new concepts introduced in the API like new types of metrics which require representation of the protocol. I would say we are pretty close. Yeah. Do we have some criteria for when we declare like that spec is being final? I'm just asking you, someone who doesn't sit on the spec seek. So, Tigran, couple of things here even though we may add new aggregations or new things that will be backwards compatible, correct? So in terms of the protocol being finalized we haven't made any commitment that will not add new features or new things but what we'll do, we'll do it in a backwards compatible way. So I think from that regard I would expect that the protocol to be very, very soon done. Yeah. Yeah. I would only add like since Bogdan you closest to metrics, open metrics project I really hope we can review our protocol before we say in the final test. Yeah. Tigran, I haven't looked at it recently but does the protocol currently contain a place to put resources? Right? Yep. It does. Okay. Just double checking that that's not a thing we're so waiting on. Cool. Yeah. I don't anticipate. I personally don't see any spec changes on the horizon except for maybe some final cleanup with metrics. I don't see anyone I've talked to proposing something that would really change the protocol. But that's the only worry that that final cleanup may remove some concepts, may redefine concepts. Yeah. Yeah. Which has a direct connection to how they are represented in the protocol or may have a connection. That is the reason. Yeah. We can discuss separately. I don't think that should block us but yes. Well, I mean, it's going to block us just saying that everything's final and interfaces are stable which at the moment it's not true. So that's okay. I mean, in our write-up on specifications repository, protocol was in the v0.4. So we were saying like let's do 0.3 and like lock this down. Got it. That's what probably Tigran is referring to. Can I go next? Because I probably need to drop off soon. Sure. I think a big focus for us right now is to do metrics. We are making progress there and we're trying to understand the specifications and it demands them as close as we can. So if you have any spare cycles to look at metrics API specifically, you're most welcome. Tracing API we're wrapping up. We're trying out scenarios and making sure that customers can work with the API. And we hope that we'll close on like all the concept for little point pieces on the VR. Next year for stabilization efforts, we probably will have a lot of rearrangements. I mean, maybe I'm not sure how big or small to make sure that all open telemetry concepts works nicely with .NET concepts. So maybe we can move some of those down to the .NET itself and that will require some changes. Those are good changes and I think we can accommodate them by at least plans and by March. So we pretty much on track I think. Excellent. All right, I'm gonna move back to the top. So we already do collectors, now we're on specs. Yeah, so specs I would say are moving along. I've got this context prop that's in its final stages. I actually was able to figure out some simplifications that Bogdan I think will make you happy that I wanna put in there. And I'd like to catch up with you in general on this subject, but that OTEP needs a couple final approvals. And once we get those, we'll be able to quickly turn this into a spec because we've already got people actively working on implementations in five languages. So I think once it's approved and people are happy with the prototypes, adding it to the spec will be easy, but if I could get an approval or final comments there, I'd appreciate it. In general, the spec and OTEP backlogs are in need of triage. They're not like too, too terrible, but I do think there's a lot of issues that are just sort of hanging around open. And it would be great to just kind of burn down the backlog. In general, I wonder if there's a way we could add some process if it's just regular triage meetings or something like that, because it seems like the specs and the OTEPs are a place where there'll be like a wave of energy and people paying attention to it, but sustaining that attention seems to be a little difficult right now. Yeah, so that's my feedback about where the specs are currently at. Do we, is it worth setting up like just like a meeting of the, well, specs say meets every week, right? Like just dedicating one of those to just clearing through the backlog? Yeah, I've been trying to add sort of triage parties to the calendar. Ended up having come down to SF this week and I kind of drew a truck through my schedule, but like next week, I'm happy to add more. I will say the spec meetings, we always seem to have enough stuff happening that we use that whole spec meeting for other purposes. So it hasn't been feasible to have it do double-db. Would it be worth setting up a special one then? Okay. Yeah. So that can be an AI. Now on to Java. Vagan, do you want to be the person who speaks that or should be the person? Yeah, we didn't make too much progress, mostly because Coupicon, I had some emergency problem, so I had to leave for a week and a half and yeah, it was a very bad time. But we should be back on track in two or three weeks. Okay. We have an auto instrumentation. If we have anyone on this call who's from there, but if we do, can you please speak up? I guess in theory I am, but I've been attending a bunch of conferences. Trask has been making progress. I've been reviewing some PRs. John Watson from New Relic has been reviewing some PRs, but I think Trask gets most of the credit, so I can't really remember for him. Okay. Seems like it's coming along though. Yeah, it's on track. I think it took a little bit of time for everyone to find alignment on exactly what we're building, but at the last meeting, Trask presented a couple of alternate designs and it felt like there is general consensus on where that project was going, so it feels good. All right, next is Go. And someone already filled this in. So it looks like Metrics SDK is in good shape. Contact propagation is work in progress and the Auto Collector Explorer will be available in 0.4 after the protocol specs version. That makes sense. Okay, glad to hear about the Metrics SDK. Next is JavaScript, also filled in. So they've completed getting started guide. They have the Collector Exporter and MySQL and Redis plugins. Wow. And a few more bug fixes, for example. Excellent. Python, v3 is released, but some features are moved to v4, including Metrics Aggregation, Prometheus Exporter. Okay, moving to an additional maintainer. Excellent. All kinds of community. And there's some work that's blocked on specs. Chris, because I know you're on the call and you're on Python. Is this work that's blocked on other languages or is it specific to Python? Yeah, it's probably blocked everywhere, except for the languages where the prototypes are being written. So, for example... Okay, that makes sense. Okay, so Metrics are still blocked on Python? Yeah, well, this is like up to us to actually spend the time reviewing the specs and the Otefs. Same deal for contact propagation. Specs are kind of being starved as people are working in the clients. Okay. Yeah. Contacts probably just by the way, Chris, we've got some people here who are gonna help Yusuke get that over the finish line. Awesome. Okay. Python auto-insertation, cool. We've an auto-installer ready, waiting on some legal ease. Yeah. And waiting on legal item. Yeah, so there is an auto-installer. We've been helping port it over based on the Datadog flavor, but because that had a different license from the Apache one and yada yada, we've just been waiting for the lawyers at Datadog to be satisfied that everybody's happy and we're dotting the i's and crossing the t's. Okay. Yeah. Next is Ruby. Do we have anyone from the Ruby work group on the call? Okay, I'll assign that to Francis. C++ update focuses currently on metrics, would like to review, like review of the metrics API that's currently under development. I don't know who wrote that, but if you wanna speak up or add your name there, I'm sure you can review the spec changes. Oh, no, you're asking, sorry, I think they're asking for a review of the metrics API by someone else. Is that? Yeah. Okay. Dot net, Sergei already did that. Now on to Rust. Do we have anyone from Rust on the call? Thanks. I know I've not been directly involved with the Rust group, but I do know that that is now a real thing. They have ported over an implementation based on, from what I understand, I'm sorry if I'm getting this wrong, but based on a pre-existing Rust project meeting up with basically the Go API. So looking at what was made in Go and making a Rust API, somewhat based on the Go flavor of open telemetry. But there's some brave souls have already deployed it to production from what I've heard. Yeah. Yeah. So, yeah, they're hopping along. Okay, awesome. Anyone from Erlang? It's okay, a lot of Tristan. Website, I don't know if Austin's on the call. I was out last week, so I had no idea if anything was done on the website. I kind of doubt it, just because, oh, Amelia, you can speak to it. I can, Austin is out this week. Oh, okay. So probably nothing was done on the site. That's fine, we need a big push before KubeCon, so that makes sense. Yeah, as always, just a general request if people are interested in, as SIGs are moving forward, we still need basic content on the website. So if you're interested in writing, whether it's the overview content or something specific about the language you work in, definitely anything is better than nothing and there's a lot of nothing on the website right now. So please help. And we're more than happy to proofread. And finally, education, not formally SIGA. I didn't write that, but if anyone has context, could you please share? Oh, I think this, probably I think Liz put this up last week and I just kind of blindly copy and pasted it over, but I do think, yeah, we're gonna try to do some workshops at KubeCon, EU. I'm not sure if anyone else has any other workshops or meetups or things planned, but if you do, please list them here. All right, that's it. Are there any other topics for today's monthly meeting that we wanna go over? Ooh, actually I do have one. We were talking, we talked about this offline, but moving the community meetings to not be in the same week as the public governance meeting, and so there'll be two weeks off. So that way then there's two sets of opportunities for the community to actually reach people, a wider set of people instead of it being Wednesday afternoon to Thursday morning. Sure, and given that January is the new year, maybe that's a good opportunity to do that. Okay, so I'm just looking at the schedule now. Yeah. So the governance public meeting will be on the 9th of January. So are you suggesting we just move the monthly no team of the community meetings maybe back by one week or two weeks? I'd say two weeks, that way it's every two weeks we have something. Makes sense. All right. Is it okay if I just make that change? Yeah. I think it's... Bogdan, I can't hear you, sorry. I think the rule is the governance meeting is the second week, correct? Looks like it, yeah. And we can make it this the fourth week. Yeah, yeah, monthly on the second Thursday for the governance public meeting. And this can be monthly on the fourth. Yeah, okay. All right, I will make that change and I will update the calendar and I'll send out a message on Gitter to everybody. Yeah. In FYI, it might be slightly weird for Morgan when you look at it because the monthly meeting, community meeting, also alternates... Yeah, I'll update it twice. Yes, I know this. Okay, cool. Yep, all righty. And I'll add that to notes right here. All right, any other topics people wanna bring up? There's fun seeing everyone at KubeCon. We should do more regular meetups. Yeah. Cool. I will see everyone when I see you. If you don't have happened to attend any OT meetings prior to the new year, happy holidays and have a happy new year. Yeah. Thank you very much. Hello, hi. Hi. Hi.