 Shall we start at the top with Schema Hero? Does anyone want to correct my pronunciation? This is basically a single person project at the moment, which I think is the first single person project we've already seen, which is interesting. And is that person, I can see replicated.com associated. Yeah, there are a bunch of commuters from the replicated to this project. Wow, okay. There's one person with almost every commit with 362 commits, and then there's a couple of other people with a very small number of commits. It's 300 and then 20. I don't necessarily see that as a problem, given that there's evidence of collaboration. If it was truly one person and in a personal repo, then I might definitely say it's a problem. No, I mean, I think that coming here to get more people to contribute is a worthy way of using Sandbox particularly as well. Yeah, and then all the other thing I would say is in a lot of projects, a lot of projects are one or two commit are heavy, especially in the beginning, because they're the primary developers and that's just how they're, that's how they're rolling. So yeah, anyway, it meets the sniff test to me anyway. Do we know anything about, they seem to have all the boxes ticked, I think that they, and they seem to be, I can't like this. Why do you want to contribute your project to the CNCF question? Seems like a completely decent response to that. Yeah, I mean, it seems like all the right things. I mean, I suspect there's a startup behind this thing somewhere, but it's replicated.com or whatever. It's cool. Yeah, they have like an installer product. So I don't think this is something they necessarily sell. It's just part of their solution. Very good. Anyone got any more comments or shall we do a vote? Let's do it. So for those who haven't done this before, we just do it in the chat and you can type your plus one or minus one or whatever you want to put. Next one is cloud development kit for Kubernetes, CDK. This comes from AWS by the looks of it. It has a very clear scope. It's like Pulumi for Bruneris. Good support. Has anyone looked at the relationship with CDK and whether it's independent? I didn't take a look at it in detail, but that was my kind of curious question. Is it independent of CDK that it can go in its own direction as a project without being because CDK is not a CNCF project? Is CDK AWS? Yes. It appears to me at first glance it appears to be independent, although it's a little bit tricky to know deeply, but certainly nothing in the docs even on the GitHub page seems to indicate that there's a tight relationship there. And then saying, hang on there, I see if they're hanging. It is based on design concepts and technologies behind the AWS cloud development kit and it can interoperate. So that sounds like we'll be using it as a library-ish. It sounds like he's in some kind of really noisy machinery. Sounds like a lawnmower or something. It's my fan. I think it's a fan on my computer. Sorry, I'll switch. Okay, that's awesome. That is some fan. Liz, you're muted in the nature hell. Oh, it was my fault. There you go. I thought if I was going to criticize Brendan's fan noise, I thought I should make sure I was looking after my own. Yeah, so any other thoughts or comments or concerns about CDK? This seems like a pretty interesting idea. All right, should we move to a vote on it? Am I better now? Yes. A little better. A little or a lot? It's a lot better, but not perfect. Yeah, lawnmower is quiet. But you can still actually hear something? You can still hear something? It's kind of like a light background noise. I wonder if it's my laptop. All right, sorry. CDK, it's done. Athens is up next. This is from Yahoo. Am I the only one who don't see the alignment with the existing open source, CNCF and open source privates? Is that column empty? I guess it's optional. Now I see something. I see as we move to the zero trust network model. Column J, right? Oh, yeah, right. Oh, there's another one. Yeah, it's very similar, but that's all right. I mean, they've listed spiffy inspire. As similar projects, it would be interesting to hear what this does that spiffy inspire don't do. I mean, maybe the answer is there is a huge amount of overlap, but they have been using Athens for a while. Looks like the repository was created four years ago, but probably private. I think it's been used for a long time, as far as I know. I mean, it talks about Kubernetes integration. Probably one of the least Kubernetes, it's not necessarily a problem. It's just being one of the least Kubernetes native projects on the list. It's just been around for a long time. Would we like to understand more what differs the project from spiff inspire, the angle? Yeah, I'm not sure the differences yet. I mean, they both use 519 certificates authentication. I don't think it's a bad thing if they overlap, but it'd be really nice to have some more of an understanding. Yeah, just curious. I think I agree. While we have the multiple projects, it's fine. Multiple solutions to a problem is fine, but I think we want to, I think we should have some understanding of what the strengths and weaknesses are of potentially competing projects. Yeah, and I also think these kind of discussions almost encourage them, forces them to clarify their positioning. And it actually helps in adoption once they become sandbox. So at least folks coming in from a Kubernetes background and CNCF background can better understand why they should check out this project. I'm also really curious if there is some sort of spec they plan on creating or are following. It's nice that Spiffy has this spec. I'm curious. Yes, I agree. It would be interesting to understand if there is any, I'm going to guess there's no relationship with that spec because they seem to predate Spiffy, but yeah, it would be good to understand that. I think I'm also intrigued by the RBAC that they have, whether that's integrated with like Kubernetes RBAC or a whole other mechanism, not that it has to be. It supports lots of non-Kubernetes runtimes, so it doesn't seem as it is, I mean. Can you hear me now? Yes. Cool. All kinds of auto and kind of all kinds of audio challenges this morning. I also noticed it looks like maybe they have a little bit of overlap with cube to IM and it looks like there's some specific stuff around AWS, which isn't bad necessarily, but it seems like it may be, it feels a little bit like it's a sort of pre-cut solution that they're looking to put out there rather than something that's being incubated at some level, and I feel like that's a little, I don't know, it feels like it might be a little odd to put something that's fully formed into Sandbox. Because if the goal for Sandbox is like, you just want to put it in the neutral IP space and you want to kind of throw it in a place where other people can collaborate, is it bad that it's mature if that's all they really want from it? I mean, it's not terrible, but at some level like if it's not, like I worry about it being a dumping ground, where, I mean maybe another way of saying that is like, I wonder what the plan forward is, right? Like is this something that they're actually using or is this something that, what's the motivation? Like I guess in an incubating project it's very clear to me what the motivation of putting it in the neutral IP is, which is build, grow community. This looks like something that's in production usage by Yahoo. And so given that, I wonder like what, what are they hoping to get out of the CNCF? Fair question. I mean they've answered the question by saying they want to work with others in the community to evolve Athens in sort of a normal way. So it sounds on the face of it as though it's being contributed to, but I do hear what you're saying about sort of mature projects. I mean, maybe they want to contribute it so they can get more integration with things like Spiffy. Yeah, maybe they want to evolve it to be more cloud native and this is a forcing function for them. Yeah, but I will say that I've also seen, and I don't mean that this is an example of this, but I've also seen projects that have, that have been defunded that people kick out into open source as a way of keeping them alive within companies. So I just want to make sure that that's not what's happening. I mean, and I don't necessarily think it is what's happening here, but I, and maybe actually we need to rephrase the question or, you know, like some question around like, oh no, there is the why do you want to, never mind. There is the why do you want to contribute. So I thought we had to ask that question, but never mind. And it looks like they do, I mean, they do state that they want to evolve the project. So I guess that's enough for me to, or I don't know, is that enough for us to kind of trust like, hey, like that's going to happen. It looks like a pretty active project, just looking at the code frequency graph on Yeah, and it does actually say used at scale. So it sounds like it is actually still used inside of Yahoo! Horizon. So it's not going to get dumped. And Yahoo! was like super early to containers and stuff. So it'd just be interesting to have that type of a company project into CNCF. That's nice. Yeah, I mean, I guess I don't see a lot of damage coming from putting it in CNCF. So in that sense, like me as well. Right, like, it's not like they're going to go monetize this thing, you know, or like, you know, start touting that people should adopt their sandbox project or whatever. Do we want to ask them as Chris suggested in the chat, we could ask them for a bit more information on their kind of spiffy spire differences. Yeah, that'd be nice. Just to understand the project a little bit more. All right, shall we do that and not hold a vote on that one? Anyone object? We move on to the next one. And the next one is CERT manager. Woo! Pop soap in a champagne bottle. Plus one for me. I don't know if that was obvious. It's in here. Would anyone like to make any comments or discuss or should we go to a vote? Think about it. Just for posture. Just, I don't even think there is a discussion necessary, just for posture. It's really nice to see CERT manager in sandbox and so many CNCF projects and ecosystem projects and people in our ecosystem like rely on this tool. So, longer overdue. Yeah, really agree. It's great. Okay, the next one is if the next one is too big to fit on my enormous screen, how does that happen? I think it's open cruise. So, this is the one we talked about before, right? That's correct. This is a reapply. And this is the one that we asked the Kubernetes steering committee how they felt about it and they seem quite keen because I think last time we discussed it, we were wondering whether it made more sense as a Kubernetes subproject and the steering committee said no, we think it stands alone as a separate project. I think I also had a question around the scope of this project and where the boundaries are because it just seems like it's a lot of components. Yeah, I think I'm trying to remember where I saw. I'm pretty sure I did see an answer to that because it was a discussion around like is this a home for all CRDs and the answer was no. And I think in other respects, it had all the right links and bits and pieces, didn't it? That's great. Yeah, it looks they presented to SIGAP and SIG architecture and SIGAP delivery. And the response on the why do you want to contribute question, I think is very good. It's very cognizant of people being not necessarily able to adopt it if it's associated with one big company like Alibaba. Any other comments or should we move to a vote on that one? Okay. If I'm driving my spreadsheet right, I think the next one is Tinkerbell. Talk about this at some point. They presented to the C Grand Time about a month ago. I know I've seen this name before. It is. Similarity with Metal Cubed and a couple of other options as well. Great name. Any comments or thoughts or should we do a vote on Tinkerbell? It's a great project. That's a very interesting news case of a bare metal provisioning. How does it compare to Metal Cubed? I guess to some extent it doesn't matter too much. This implementation is different. I believe Metal Cubed uses, I forgot the name of that open stack library. Yeah. I think the goals are similar. Metal Cubed uses, was that? Triple O? Oh, no. Someone opens. Ironic is the name of it. Ironic. Metal Cubed also has integration with plus three PI of Kubernetes. That is implementation details, but the goals are similar. Tinkerbell, I think it came from Packet, which was a fantastic service. I used Packet a lot. I never used objecting isolation, but I suspect it's pretty good. Shall we do votes? And then Stanza. This is interesting. It's a replacement for Fluent Day and Fluent Bit. It's a very new project. I think it was created in July this year. It also talks about aligning with open telemetry. It does currently have 23 stars in GitHub, which does suggest it's pretty, yeah, hasn't set the world on fire. I mean, it was created in July this year. They're just so far, they're three contributors, three active contributors. Yeah. But the claim is modern replacement for other projects. And you're just a three-month-old project. I mean, I don't know if we should be concerned. Yeah, I feel like we should push back a little maybe and just ask for, I mean, I hate to ask for sort of a min bar, but it does feel a little bit like it's not, like it doesn't have a ton of momentum there right now. There's a few places in their application where they talk about interest within open telemetry. And I would have expected that interest to kind of, like one of the artifacts of that interest would be a lot more stars than 23. It seems like one of the major selling point is it's written in Go. Yeah. It's a lovely programming language, but I don't know that. I mean, I think that's one of the reasons why Fluent Bit is perhaps more, you know, has some advantages over Fluent D because it's, I think Fluent D is Python or Ruby or something where it's Fluent Bit is C. So there's a performance implication there. Yeah, I feel like if only they just said it's a Fluent Bit, a replacement written in Go that people would be able to understand it better. All this other stuff, it's really hard to quantify. I think that is largely what they are saying, but it's... I know, but it's buried in the middle. Yeah. I think that's the real differentiator. So if you, I mean, if someone asks me if there's a good, you know, replacement written in Go, is that worth a try? I'd say yes. But again, I have no idea how given this is such a new project. Yeah. Yeah. It's not obvious that they need to be in the CNCF to drum up more interest in this. Yeah. I would also say that like... Oh, sorry. Yeah, I just noticed that there's no issues open by people who aren't contributors, which suggests that no one's trying it out. I was also going to observe that their project boards, I was looking at their project boards, their project boards are a little thin in terms of road map. I mean, there's some stuff that like... And it's the same sort of issue, but it's like 10-ish kinds of projects on the next release. It's just, it doesn't feel super... Yeah. And in fact, most of them are all written by one person. So like it just, it feels a little bit early. I think what we should maybe communicate back is like, directionally things make sense, cloud-native makes sense. But it's just super early and you need to spend another six months drumming up interest and usage. Yeah, I would have a concern. I think that perhaps this is a startup hoping to generate marketing buzz around themselves by having a project accepted. I mean, that said, I think the goal, I mean, either Fluent or Logstash, like they're both kind of bloaty programs. So like, and written in Ruby. So like it's not a bad thing to replace it, but yeah, it feels a little bit early. Yeah. Do we want to vote on it for now? Or perhaps I should put it another way. Does anybody want to hold a vote to say? I feel like it's, we don't need to tell them we rejected it. I think we just say we think it needs more time. I wish there was a way in the vote to be sort of like zero. Can I vote zero? Yeah, I feel zero about it. Like I don't want to vote negative one because I'm not like down on it. I just think that it needs more time. Agreed. More time, more interest, more. Yeah. All right. Yes, let's ask them to reapply in future, Amy. I think we have some guidelines somewhere about six months or something. We do. And I can point to it. All good. Great. Pravega, is Pravega, Pravega? Pravega. So they were, they applied for incubation, but and I reviewed them and basically told them they didn't have enough users and users. And so they decided to apply for sandbox. So, which I thought they weren't going to do for a bit. And then they decided they were, they were happy with that. So I think, I mean, they're, they're trying to compete with Kafka, but they, they really need to use a momentum. But the project's been around and it's relatively, what's here as a project is used internally and by kind of customers, but not much externally. They've just launched a rust version as well, just a couple of days ago. This is, it was pretty Java based and doing some rust work as well. I think the storage, six storage took a look and then. Yeah. So they, they, they recommended it and they're interested in doing much more integration with Kubernetes and things. So that's why they very much, it's about Kubernetes oriented projects and their view and they, that's why they want to be in CNCF. Any other comments or thoughts before we do a vote for Pravega, Pravega, sorry, I'll, I'll learn to say it right. All right, Kaiverno, I wonder whether this is a project in its own right or is it uses, does it use Rego? No. No, in fact, they actually explicitly, explicitly say that, that they're, one of their values is not using Rego. Yeah, it's an alternative. And we, I think we've, we've had people saying that they would like to explore other options for policy. So kind of, it seems quite, Jamal seems to be very focused on specifying policies in me, but I guess I'm on the, I like Rego side of the world personally. Yeah. It, it seems to have, you know, above the bar participation and interest as well. So, you know, whether, whether it succeeds or not is not really for us to judge, but, but it does seem to be, you know, active issue. It doesn't have a ton of contributors, but it has more than say, like the last one we looked at. And it has, well, Does it have more than? Oh, sorry, I meant the one before that more than the, the one that we kicked back to people. Oh, right. Yeah. Yeah. Okay. The stanza, stanza. Yeah. So it has reasonable activity. They've been able to garner some interest. There's 50 forks. I mean, so they, they garnered some interest somewhere along the line. I have come across them in the like Kubernetes sake security, I think they're involved in. I don't see what looks like a lot of like user issues in their, well, maybe a few in their issue queue. Not that that necessarily matters, but as I look through their issue queue, it's mostly, it's mostly a project board rather than, than end user issues. But yeah, it's early days. So it doesn't really matter. Yeah. I think personally, I'm having a little bit of trouble dissociating. Yeah. The kind of the rego question from like actually, you know, they seem to be going about things the right kind of a way. Maybe they're going to create themselves right through writing policies in this way. I'm, yeah. I kind of wonder how you really combine policies and things that I mean, what the, I mean, does it have a, does it have a model that you can not make mistakes in? Because that was, I mean, there's some issues with that that we were raised on the security evaluation for Oprah as well. That feels like a little bit of a high bar for sandbox though. Yeah. No, completely. I think, yeah. I think the sandbox is probably a good place to look at those things rather than, yeah, it certainly seems to me like they're doing all the right things and they have, you know, it's not like they're lighting the world on fire, but, but not like it's three people in their garage either. Yeah. And I think, I think diversity in the space is good if only to push Oprah to do better, right? True. So I think that I like that as well. They're also, you know, decent. They're associated with Nomada it looks like. So it looks to me like they're not going to like super CNCF, hey look where sandbox about this project either, right? It feels like it's going to be a little more open than just, hey, our startup needs this thing to go. Yeah. I think I agree with you. They look like they're, they're wanting it to move into a neutral space for good reasons. Yeah. And it's not the only product in Nomada. Like if it will be a part of the thing, it's not the only thing. And so I don't, I don't see them like getting too noisy about being in sandbox. Better not. They incur the rough. That's right. They need a better logo though, if I'm being honest. That's definitely something that can be fixed in sandbox. I was going to say, is that, oh God, yeah. All right. Should we, should we do some folks for Kyberno? And the last one on the list is data set lifecycle framework. Not sure. I'm satisfied with the answer that they gave for differentiation from the existing cozy project that's underway. And I think the answer appears to be as soon as that becomes viable, where we'll stop this project. This is their answer in Coloman. Is that what you're saying? I believe it was, let me see. Yes, Coloman. It's cozy, going to be, going to be mature soon. I'm actually not familiar with that. They've gotten kept approval and they're working towards getting a POC completed by the Kubernetes 120 timeframe and then an alpha and the 121 timeframe. Oh, so it's part of upstream Kubernetes, I see. Right, right. I mean, I must say this, when I look at this project, it looks kind of thin because you know, it's cooling some way because it's using Kubernetes as an app platform with the database to kind of encapsulate all these other concepts. So that's actually kind of cool. But if it's going to be, but given NFS and S3 are such fundamental concepts in infrastructure, if it's going to be captured as part of Kubernetes storage API, then it's really, we should probably at least wait till see how that thing plays out. And this project still seems pretty thin right now. That's my impression. Yeah, it's another one with a fairly small looking level of excitement to judge by. There is a release valve. If you wanted to push them to another organization in the Linux Foundation, we do have this LFAI and data organization, which is kind of all data people, AI people. So this could be a better fit there, but it's your call. Well, the other thing I did hear a little bit about is it looks like it might be a research project. I think it's that IBM research. Yeah, it's from IBM research. Yeah, it looks like they got a grant. Like at the very bottom, it was funded by the EU. So worry a little bit about it being more of an academic project than a legit end user facing product, but I don't know. The FAQ links kind of three people, but that not two of them are just kind of interested or close to similar proposals. And only one of them is running a POC. So it just seems that they are very far from having actual users. I mean, maybe the right answer here is interpreting that sentence that Saad highlighted before about when the cozy interface becomes part of Kubernetes, we will stop maintaining and directly support cozy. I don't, they say we will stop maintaining the whole project or they say they'll stop maintaining. So there's a little bit more in the FAQ where they say when the cozy interface, we will stop maintaining and directly support cozy for creating PVC for buckets in object stores. So I think what they mean is the framework stays, but instead of talking to CSI directly, the CSI plug-in directly, they're going to talk to the cozy interface. And I don't know very much about it. So if I didn't make any sense, Saad should tell me I'm not making any sense. Yeah, I'm honestly not sure. Okay. Then what else is left? I thought all this is is kind of a, you know, is a YAML interface to S3 and NFS provisioning. That's what it feels like to me. I think it's automatic PVCs. It sounded like, right? Like you just pointed at a bucket and automatically a whole bunch of PVCs are created, but I'm not sure. So you can say dataset basically, you can say, hey, you know, like, if, and I can imagine like if I'm a scientist and I don't know anything about volumes, but I know that I uploaded a whole bunch of, you know, data files. I just want to point out that's something to be like, make, make go, you know, it sounds like that's what they're aiming at. So it runs an NFS gateway on S3? No, that's no, no, it like introspects. So it goes into introspects S3 or it goes into introspects of the NFS. And then it just creates a whole bunch of volumes. At least that's my understanding. How does, how does S3 turn into a Kubernetes volume? I'm confused. I think they use CSI for that. No, but how do they, I mean, doesn't Kubernetes volume has to be a block device or a file system? So how does S3 turn into a block device or file system? Well, I don't know. There's an S3 based CSI. So I assume that it does something. I don't know the details, but in the fact, you can see that there's this. Is that maybe there's a CSI implementation for S3? I mean, you can, it could be using Fuse. I don't know. It could be using like one of these HTTPS. It says that it's using Fuse. Yeah. I mean, Fuse S3 is a thing. Yeah, that's right. That's probably it. So it's basically running a gateway called S3. And it sounds like effectively this is a way that you could sort of create a meta set of volumes based on some storage back end. I don't know. It's a pretty thin use case in my mind, but I kind of get it, I think. Like if you're a scientist and all you want to do is run an R. So does Cozy cover this as well? Like turning? No. So Cozy's focus is not taking object and turning into a block or file. It's just focused on surfacing object directly into the containers in kind of a standardized way so that it becomes easy to consume. Actually CSI S3 sounds actually sounds actually kind of interesting. CSI S3 is basically just a Fuse driver in front of an object store. And so from what I can tell, this is kind of taking all of those use cases and trying to make them easy to use, I guess. Yeah, I would suggest looking at the cube flow, like in the table of contents over on the right hand side, there's this like model storing and surveying the DLF. I think that would give you a better, gives you a better example of what they are, how they're thinking about it. What do you all think of Chris's suggestion about the LFAI Data Foundation? Personally, I think I would kick them back for six months and just say, hey, talk to us in six months. Because if they're dead, because like there's multiple questions about how they evolve, and I feel like there'll be a lot more clarity. I mean, to be fair, if we think that they are genuinely kind of cloud native, you know, I'm not quite sure how to articulate this. No, I think Liz, I don't think this is a AI data kind of thing. I mean, these cases we're talking about here are, I think, are more generic. So, you know, there's this cozy thing. And then, sorry, CSI4S3, is that just a third-party project they're using? Yeah, it's a third-party thing they're using. Yeah, so then they're just packaging. Because if they just did that, that would actually be interesting. They're not packaging. No, they're not packaging. They're making it easier for you to reference. Like you don't have to deal with VCs or anything like that. It's an admission controller, basically. It's kind of like the service mesh for your data, I guess at some level. I think it certainly has potential. It just seems quite confusing and quite early. Yeah, that's why I would say so. That service mesh for data, that's very compelling, I think. Well, I know, but that's why I can get things funded from VCs, no. I'm kidding. They can only get grants. But I think it's worth just telling them, hey, look, interesting, but there's not enough there. There you need to spend six months and see. Because if in six months nothing's changed, then it's very clear that it's dead. And if in six months they have, they say that they're a service mesh for data, then maybe it's a thing. No, this is actually one of our second applications. No, is it? I don't know if that changes your calculus at all. I am happy to be able to say reapply in future for spring review, because that is in fact an answer. Does anybody remember what we said last time? Yeah, I think we were asking them for clarification on their differentiation from COSY. Okay. In that time, the activity hasn't set the world on fire. Do we think that they are at a disadvantage by not being in the sandbox in terms of collaboration? They're an IBM project. Do we think there are potentially people who would collaborate, but are not doing so because it's associated with a particular company? I don't think so. So here's my thing about this is this looks like something that you would install on a Kubernetes cluster that's being used for data science. And in that context, nobody cares about this stuff because you're just trying to make the scientists jobs possible. This doesn't feel like something that would be adopted into a cloud where competition would be important or something like that. So I think the fact that they can't point to a bunch of people using it is an indication that maybe they're not quite nailing their use case or they're not quite nailing their PR around it. They're going to market it better service mesh for data. I'm telling you. I'm telling you that maybe Amy, that's what you should just tell them. Go spend six months at the service mesh for data. I want to cut. If anybody starts their startup with that, I want to cut. Anybody on this call? Definitely. It sounds like leaving it at reapply in future for Spring Review, but also letting them know that the LFAI group exists might be helpful. Sure. And I think we are still looking for clarification on COSY. And what they came back with was when COSY exists, we're going to deprecate a bunch of stuff. I'm not sure that's clarification. That's just saying they're going to use it. It's another reason to think, well, if they reapply, we might have a bit more understanding of this whole how it shakes out with COSY. Yeah, I think so. Also, there were seven PRs in the last month, which is pretty light. Yeah. Anybody object if we tell them? Well, like Amy said, come back again in the future. But by the way, there is also this other foundation. Okay, let's do that then. We're back on time. Kind of, yeah, close enough. Our next review will be in January if we want to be able to keep with the two-month cadence. I think we should do six months. Oh, you mean for the sandbox? Yeah, for the sandbox review. Yeah, yeah, yeah. The next time we go through the spreadsheet would be in two months, which I think is good. Yeah, with the holidays, it makes months. I might sketch towards the later in January than earlier, given because like I know people might not be back in the office and things. So I'll say we're tracking towards January. Okay. Awesome. All right, watch for an email from me with like results in here and then I'll push it out towards the main list. Good to see you all. Awesome. Thanks everyone. Bye. Bye.