 All right, we are recording. This is Tuesday, January 23rd, and this is our sandbox meeting. That's correct. All right, welcome everybody. We've got 10 projects in our upcoming queue. I don't believe that we'll be able to make it through all of them in the next 29 minutes, but let's get started. So the first application is Connect. It's a simple cross-language framework for the Protobuf RPC. For building strongly typed APIs, they currently have implementations available in Go, TypeScript, Swift, and Kotlin. Was there any TOC member that reviewed this project and has any comments? I believe this one did get a recommendation from app delivery for inclusion into CNCF. Any comments? It's an interesting one, this one, because it's, I mean, that list of languages is because it's targeted at a very different audience, I think. And the only example they gave of things similar in the CNCF was GRPC, and everything else they linked to was actually an Apache Software Foundation. So I know we had these various discussions with GRPC about, is it really a cloud native thing? I kind of asked the same question here. Is it, do we want this kind of thing? I mean, I think as a project, it looks fine, but it's clearly targeted at mobile and web developers, to a large extent. And it doesn't have any usage in CNCF that I'm aware of. Nikita? Yeah, kind of similar thoughts. Like I'm missing the connection to cloud native here. And also, I think Justin pointed out on the issue that some of the applications surrounding Apache projects are not CNCF ones. So App Delivery and App Network, both of them said that they would recommend this project for Sandbox, but I think that there is no guidance on why. So I think we could probably go back to the tags and ask why they think this would be a good fit and take you from there. Do we think that if the tags were to come back with a recommendation, because I'm kind of an agreement here, is that very similar to GRPC and some of the conversations we had around that time that they had applied for moving levels, this project to me feels slightly outside of CNCF, but still supportive of a lot of that. Thinking back through open features, adapter interviews, we had some mobile development challenges that they were experiencing, but there is that part of our end user and our adapter ecosystem that does mobile-based development. So I'm wondering whether or not we should extend the project pull the way that we have with GRPC to include this one for Cloud Native or whether or not this doesn't really kind of fit. Are there folks from App Delivery here on the call that can provide additional information around that recommendation for inclusion? Great, now I'm gonna call on you. Sorry, trying to unmute. So when they presented, so like Josh said, approved, but what Justin's saying makes sense, I can't, yeah, and what Chris said, I would lean more towards approving or CNCF sandbox, but we are having a lot of API discussions and we have an API call on March 6th to have a whole hour talking about APIs and how they fit within Tag App Delivery and invite the other tags to participate in that. So it was an FYI. So TOC, do we want to send this back to the project or do we want to move this to a vote? And if we send it back to the project, we need to be very specific with them for what we're looking for or send it back to Tag App Delivery with a more specific recommendation for inclusion. I have a suggestion. Ricardo, go ahead. So it seems like you don't know whether that fits in to Cloud Native or not and it might fit into App Delivery. So you could potentially say that you're working on changing or extending the scope of App Delivery to include these kinds of projects and maybe they can come back after that. That definition is complete or that change is done. But that's just a suggestion. I appreciate it. Unfortunately, Tag App Delivery is very overwhelmed with work, Karina. If there's not a rush on submitting this one, that's not a bad I thought because they could come participate in the API discussion and we could get more feedback from Type Network. Okay. To you see, how would you like to proceed? Yeah, I'm not hearing a desire to vote on this either, which means we need to have next steps. So since we don't have a desire to vote on this, what I'm going to recommend is the project engage with the API group and Tag App Delivery. Understand a little bit more how this fits into Cloud Native and the value add that it has to the ecosystem and then provide additional recommendation on this application so that we can bring it back to a vote after that API meeting. Does that sound something amenable to the TOC? See a thumbs up from Justin and Nikita. Okay. Is there a TOC member that would be willing to provide that comment back to the project and Tag App Delivery? Who are the liaisons for App Delivery? Justin, would you mind taking this on as your last act? Yeah, I can do that. Yeah. Thank you so much and we will make sure that that's transitioned appropriately with a new seated TOC. Thank you. Okay. Next up. Kairos. It's a metal Linux distribution to build immutable edge Kubernetes. Any TOC members review this project? Have any comments on it? Observations? I thought I recall reading that there was a recording on this from the tag. Ricardo, you had a comment. Yeah, I think it's ready for a vote. Okay. One thing that was concerning to me that the list of maintainers has all maintainers from single company. Yep, we're going to expect them to increase the maintainer diversity for company affiliation, but it's not currently a requirement for sandbox inclusion. So, fully noted Nikita, thank you. Yeah, I have gone through this project. I think it's a very good feature because it allows, you know, very easy, you know, build and deployment on the edge site. And the only thing is the project name is, you know, it's the same as a company's name. Is that okay? That will need to be a trademark discussion that the CNCF has. Yeah, we usually have that as part of like onboarding. So they could either change the name of the project or change the name of the company. I'm not concerned. We generally fix it as part of it. Okay. All right. Do we have enough information to move this project to a vote? Hold on. Sorry, just remind me. What did we decide about accepting Linux projects? I don't know that we did. Because, yeah, I mean we, because Microsoft was had what are you talking about? We had a flat car and so, and we didn't make a, my memory is we didn't make a decision. And so we're just making a decision for these ones and not making a decision for flat car. That's kind of weird. Well, that car is actually going through due diligence right now because they applied incubation. Correct. Okay. But so we've decided the Linux projects are in. I don't know that the TOC has made an official decision on whether or not Linux projects are included within R2, within the definition of cloud native. So the question for this one is whether or not Kairos, as it exists here in the application, if it meets the definition of cloud native and should be included within the CNCF based off of the information presented in the application. Our decision to include it does not necessarily meet automatic acceptance for all Linux projects into CNCF, nor does it mean that flat car will progress beyond incubation because that's an independent project that we'll have to evaluate separately. But we do have resources allocated to performing the due diligence of it. Justin, you have a lot of hesitation. It's very, it's very similar to flat car. I don't see that we would make a different decision from flat car. I mean, I was, I'm in favor of flat car. So, but I was, I think I didn't get any feeling that we had a consensus on that. It's going to say, I have it in my notes that we covered this last year as well. We discussed also this issue regarding this project very briefly. And I think that's where we left the comments asking for a review from tag runtime also in this aspect. Who provided the recommendation for inclusion further down. Exactly. But I think one of the concerns was, was, was exactly that. Do we take on distributions? Alex. A good question. TUC is this a discussion that we would like to have outside of Kairis. We can place this projects application on hold and just have a much broader conversation about Linux distributions inclusion within the cloud native landscape. Or are we comfortable moving this one to a vote and loving that decision on a project by project basis. Given that we're already doing due diligence and flat car and comfortable moving us to a word. Okay. Others. I think it might be helpful for us to have a broader conversation and document that because we see this kind of projects coming to CNC up more and more. And we need to have some kind of alignment because we cannot evaluate it. On every single project basis is it's very expensive for us from a time perspective and research perspective. So I think and everything if we accept to take Linux distributions, that means an expansion of the ownership area for tag up runtime. Sorry, tag runtime. So this is something that we need to align with the tag as well. So I think having a discussion. Aligning on that will help us to move forward. But I'm not sure. I think we can do this in in parallel. It depends on what the app or TUCs think as well. Yeah. Okay. The discussion and voting can happen in parallel. Okay. My only concern with them happening in parallels if our discussion would overturn a vote. Yeah, that might happen too. So we might have the discussion first and then decide on the project. Ricardo and then Kathy. Yeah, one question one comment about this project is that it's not a Linux distribution per se. This is a project that provides tooling. To create distros. So. Yeah. It has the similar outcome for the user. I didn't think it's, I mean, it's. We should check the recording. I think we had the exact same discussion last time. Don't miss it. Okay. I think sorry. Ricardo noted. We should very check that recording and make sure we have that available before we have that discussion. Okay. Okay. Kathy. Yeah, I'm just wondering, you know, whether we have any other parties that's similar, like it's kind of like related in Linux distribution. But we already approved either sandbox or. Equating. Yeah. We can certainly pull that together. I don't think there, there are any or if there are, it's very few, but we'll pull that into the discussion. So. Let's move this along. So what we'll do. We'll place Cairo, Kairos on Kairos on hold. For a sandbox vote until the TOC meets and discusses how we want to proceed with Linux distribution projects. I mean, we'll need to fit that into the TOC calendar. Sounds good. Okay. Next up. Atlantis. This is the Terraform pull request automation project. It allows users to create PRs against a repository to run Terraform via command line comments. And there was some discussion around whether or not this should be included within open tofu, if I remember correctly. A little bit of back and forth and it looks like we did get. That discussion from tag app delivery as well from their meeting on December 6. Was there a TOC member that is reviewed this project has any comments on this for inclusion into CNCF. Any concerns. I guess I'm still missing why there shouldn't be a sub project with open tofu. I feel like the scope is too narrow to warrant a sandbox project. But I'm willing to begin with that. Have we started an engagement with open tofu? Do we know if that's a possibility? I do see that in the comments here and I don't know what has been done since December when that comment was made to the PR. If anything. I don't believe it. Emily, if you want just to try to get more information. Yeah. If you could chase that down. I think my primary concern here. And I think we had this discussion previously is projects needing to support more than just. Vendored or commercial offerings. So I know we had a project previously. That came in like this, but since this is more about a project, I'm not quite sure how we want to proceed Katie. My comment here is. The project seems as negative side is like too narrow in scope and very specific to the PR life cycle. My question is how this would evolve further down. It seems more of a plug in kind of situation rather than a separate project. I might be missing something, but this is like the scope. I'm saying it's something that is more of a plug in and can be an add on based on what. The users or adopters would like to use. But I would be concerned from sandbox where it's actually going to go further. Because I think this is like a good mature project. It delivers very specific features. And I'm wondering what's next. So I'm not hearing that this project moves to a vote. We would like them to engage open tofu and understand whether or not they are a better fit there. If. They do come back. And they don't as Karina says. If they would like to be a CNCF sandbox project. Do we feel that the current state of the project, which I think it's, it's fairly young. I remember looking at the code base. Would be worthwhile to consider for inclusion in the CNCF at that time, or should we wait until the project is more mature. It's not young. Okay. Thanks. Since 2017. I must have been mixing it up with one of the later projects. Thank you. They have like 360. Individual contributors. Yeah. So it's pretty large. I just, I agree with Katie. I don't feel like it fits within what we would consider a project. And where we have been very specific about. Notating plugins. And then projects. And so I, I can't. I guess looking forward, I can't see where this would. Go to incubation under those. You know, specific parameters. And it's also. I always struggle when a mature project does come up for sandbox as to why they're applying to sandbox. And it seems like they would go directly into incubation. If they're that mature and have that many contributors, it doesn't even seem appropriate. And it's the same. Go ahead. The same kind of question that Aaron did, just in reading through their application, it effectively looks like the project was dormant for a while and they're worried it's going to happen again. And they think that acceptance into CNCF will give them. A strong support structure to ensure that like. If the, if you read that, why CNCF section at the top. It basically says that like it need to ensure vitality of the project. And that CNCF would provide a much needed support structure to ensure that. And. I don't think that's really what we do with, with sandbox projects. Like if this thing went dormant for years, like it says in the first sentence, we would have the conversation we had half an hour ago and vote to archive it. So it seems like what they really want is to join open tofu or some other like larger. Terraform community. Okay. Okay. So. Do we want to put it to a vote for inclusion? Or do we want to recommend the project engage open tofu? And then come back if they're not accepted so that we can make a vote then. Based off of this conversation, I'm not hearing that we're, we're likely to include them. If they were to come back. Katie, you came off mute. Yeah. Yeah. So, I think we have a discussion of the option of engaging them open tofu first and understanding there is any. Potential collaboration there. If not, I would be interested to understand why? Because it seems like more of a natural fit there. But if not, I would definitely, I wouldn't like to just kind of. Completely send them off. I'd like to understand the conversation and the implication of them collaborating with open tofu. If not, then I think that's going to be different conversation on this. I think early to make a decision of not CNCF now. Okay. Are others in agreement? See some head nods. Okay. So I needed to see member to provide a comment back to the project with that. Sentiment that we would request that they go engage open tofu. Understand whether or not they can be included as a project with underneath of open tofu. Or if they would like to reapply at a later date to make sure that they can be included as a project with open tofu and CNCF that they support open tofu in addition to terraform. There's someone willing to do that. Thank you so much for cardio. All right. We have seven minutes left. Do we think we can look at. Open elementary. See some smiles. We can start the discussion maybe. All right. Open elementary is observability and observability. So let's see what we have here. We have an extensive AI applications. See here. They see this as an extension of open telemetry. And there was some discussions with tag observability. Going back and forth about whether or not this should be included. Why is it not engaging with open telemetry. There seems to be some. I don't believe we got a recommendation. Yeah. Emily. So they have engaged with the tag. And they have also presented as a project. Now there is some fundamental. Concerns that you know both were raised from the. Tag. Discussions as well as from. Open telemetry in general. Where they are. They have three issues. One is that they have patches that they are carrying. Which. Do not comply with the spec. Of open telemetry and their semantic conventions which they are. Changing and. Have not merged right because they are not compatible with open telemetry. So the recommendation was. To actually work in the project. On the semantic conventions were group. To ensure that. You know there is compatibility. The second part is that. It's a startup coming in with you know their. Idea of providing this framework. Whereas it could be best suited within the open telemetry framework in order to within the project itself. To be able to actually maintain compatibility with observability. Guidelines and spec. Requirements as well as with. Some of the licensing requirements that. Would need to be met. They have a proprietary product called face loop. And. Again. There are anomalies in what they are saying versus what. You know they're doing on the project. So. Again there's a lot of detail that was called out in the tag. Discussions that they need to address. Before proceeding ahead. Do you see members any comments. Yeah, I tend to agree. I mean I reviewed this and it was like it seemed to be very. A very early stage when there's, you know, already kind of open issues and discussion in the upstream project. I mean it's fine to fork something as an experiment, but to focus an experiment, say you want to bring it into CNCF as a new project when. There's no consensus yet about. How to do. These types of things. It just seems like as a fork outside CNCF, that's as a discussion point and iteration that seems fine, but at this point seems just the wrong time. Yeah. I agree with Jesse. So it sounds to me, just based off of that discussion thus far, that we are not moving to a vote on this project that, that we would invite them to return after they've resolved the open comments and changes that have been noted by tag observability. Is that correct? Yeah. I think, I think. I'm not sure. I mean, I think that is maybe just waiting until there's more consensus about how to do this generally. I mean, like, I'm not sure that just, you know, I would give it six months rather than when the comments are resolved or something at least. Agreed because I think it's, it's both as Justin is pointing out maturity of some of the techniques and best practices in this area of LLM observability and also at the same time, just incompatibility with large existing specs and observability in the CNCF land. Okay. Makes sense. Ricardo, your hand is up. Yeah, sorry. I just clicked the wrong button there. Well, one thing that I would say is that a lot of people are looking for projects like that, this one specifically, and it's not necessarily super easy to find them. So there is a comment in the thread that says to engage also with the AI working group. I think that's a good idea because there will be a lot of end users listing as well and probably other projects trying to focus on the same. So maybe just trying forces, maybe recommend also reinforce the comment there to engage with the group. Okay. So we're going to invite them to reapply in six months. As part of that re application, we would like to see them engage with the AI ML working group. We would like to have them resolve the outstanding comments and observations that tag observability provided them for work that still needs to be done. Is that correct? Did I get them all right? Yes. Yep. Okay. All right. I can add this as a comment on the issue for the project and this one we'll see them again in six months. Sound good? Thank you. Yeah. All right. That's all we have time for today. Any last minute questions, observations, opinions before we close out? Is that the only sandbox meeting we've ever had where we haven't let in a project? Probably. But I think we had a good discussion and the TOC has plenty of decisions that we need to be making for how we review projects moving forward for their inclusion in CNCF. All right. Well, Thank you. Thank you very much to our outgoing TOC members. Sorry, Emily, back to you. You beat me to it. I was going to say thank you so much to our outgoing TOC members. We appreciate your service and we hope to see you back again in the cloud native ecosystem. Right. Thanks everyone. Enjoy the rest of your day. Thank you. Thank you. Take care everyone.