 Can you see my screen? Yep. OK. OK, so based on last meeting's review, we're saying we want to have a mandatory section with a small number of questions so people can go over them quickly. So right now, there are 12. As I say, it's maybe 12 minutes. But I'm actually not sure if it takes more because people may need to think. So here are the questions. And how do you classify your experience with cloud native storage? And the fourth question, rank the attributes of storage systems based on how important they are. So here are the attributes that we discussed about in the white paper. So basically, just assuming when they are ranking this, they may need to think a lot. So I don't know if one minute is enough for that. But at least they can rank. They don't have to type a lot unless they have to specify some other attribute. Alex has a comment here. He said, OK, we want to add deployment options on-prem or cloud. So I guess we can add that. So the only thing is, I think if they answer what type of storage they are using, then maybe we will know already how they are deployed. But I think we can add that question, I guess. Anybody has any comment on this one? So OK, to add this, deployment option with on-prem or cloud. Where was the proposal to add to which question? You see my screen? I see it. OK, so Alex says, he was saying that we want to add a deployment options on-prem or cloud to cover. So this is in the white paper, basically talking about this one. Basically here, we're talking about you deploy your storage either as a hardware solution or data center. This is really the enterprise storage. What I didn't understand was, is deployment option and attribute doesn't seem like it fits as an attribute? No, I think it's going to be. I would think that it would be a separate question. I don't know why he highlighted number four. It's a separate question. Does it make sense to add a separate question, say whether it's deploying on-prem or cloud? I think it's interesting knowledge to know, but I'm not sure that it makes a difference to the technical need, right? If we know what attributes they're looking for and what types of storage they're wanting to connect to, where it is, whether it's in a public or private cloud, I'm not sure how much difference that would make. We probably will tell already. Because there is a question below, basically asking you to tell us what storage system you're using specifically. So from that, we probably can tell. So maybe this is already covered. OK, so I will just ask some comments here saying that. Yeah, I think if we were to ask the question, it would certainly be its own question, I would suggest. OK, yeah, OK. So I will just ask him if it's OK. Maybe it's already covered anyway. So all right. And then number five, basically just rank those attributes. It's the same attributes, just rank them. What are the pain points when you're using continuous storage or the pain points you have? Rank what is the most, the biggest problem that you have? OK, so any comments, number four and five? Yeah, so if we get them to rank the pain points, which I think is a great idea, does it also answer question number four? I think that number four is what is the most important. So basically, you first to say, OK, when you make decisions on choosing which storage system to use, what are the most important attributes? So you look at those. This is a little bit different there. Yeah, yeah. What are they their choice on? OK. Right, the second one is really when you are using them, then, what are the problems you're facing, because it's slightly different. Yeah, OK. Yeah, so then number six, basically just if you are using block stories, tell us what is the system that you're using. So if they give us a name, then we'll probably have a good idea, right? It's out there as Google Cloud, we know. Deployment is on cloud, not on prem. And then if you're using shared file systems, specify what systems you're using is the same. And then we have the same questions for object store. Just tell us which one you're using, key value store. So we give several choices here. And then if it's a database, what is the database you're using? So I think there's some overlapping between number nine and number 10, key value. But some of those like Cassandra, Cochrochi, those are also databases. But I think they can also do key value. So that's why they are kind of in both places. So if it's not already in number nine, then specify that in number 10. Yeah, I think that's appropriate. Yeah. OK. And then the next one is the container orchestration systems. What are you using? And then after you specify that one, what type of... So this one could also be multi-slap, I guess. Most of those are multi-slap. And what type of workloads you are currently running on containers or what are you planning to run in the next six months? So we'll give a few examples so then you can list whatever they are running. I think based on your examples, you're asking for stateful workloads. So maybe you ought to put that word there. Otherwise you could get anything. Oh, OK, so like this? Yeah. OK, so that kind of, yeah, sure. Otherwise maybe it would be all over the place. OK, so here are the mandatory questions. And then we go to the optional ones. So this is the how is your storage solution deployed. So do we need to combine this with the comments from Alex saying on-prem or cloud? Or do we still keep this? This one basically saying, OK, you deploy that as hardware solution or software components on commodity servers. Yeah, this one still has a little bit more detail, I think, public cloud or cloud storage as a service. That is, you have some like Cassandra as a service, but you offer that on a public cloud. So that's a little different. So I think this one still has a little bit more detail compared to the one that Alex mentioned, that's on-prem or cloud. That's kind of more general. So I think. Yeah, I'm not sure if it helps us to know the difference, but one difference might be in your local data center, right? A data center you're managing or in a data center where you are co-located. OK, so it's a public. OK, so it's on-prem or cloud then. We'll have to add that. We can add then the separate question there. Well, no, I'm sorry, if you see the first option here, how is your storage solution deployed, deployed hardware storage solution in a data center? Oh, what type? Was it private or public? You might have one option being in a local data center. And the next option might say deployed as a hardware storage solution in a co-located remote data center. Co-located data center. Right, so. So co-located data center, what is it? Is it between public and private? Or what is co-located? How is that different from public? That's not public cloud. I would not consider co-location a public cloud. You could lease the space in the data center. Those machines are your machines and managed by you. It's not a public cloud. I think if they also answered another question about whether they were on-prem or in cloud or co-located, we could probably deduce this. Yeah, if there's a co-location somewhere else, I think that those three are the three that we'd like to know as far as the comment made further up in the form as to how are they deploying. Those would be, I think, the deployment options that I would consider of interest. OK, so should we? The only thing is, at this point, hitting question 13, I think we have to maybe go back and decide which of these to call, because I'm hoping that to get participation, you want this to go out with a promise that would you please take this 10-minute survey? Perfect. In 10 minutes, you're going to start to get a lot of people who are never going to even open it. Right, right. So those are quite optional, right? So right now, we are in optional questions, so we can perhaps a little bit more freedom to do here. Yeah, well, I think Stephen's point is we don't want to ask the same question in a variety of different ways throughout the library, right? Yeah, exactly. OK, so I don't think we have explicitly, and we may want to just kind of peruse. I'll have to open the document in another page here. And we can just look through it to see if we are asking the same thing in two different ways. Yeah, so basically, can we just look at those two questions? Because we didn't really ask the deployment questions before, other than we just ask them what storage systems they're using, then that we can probably tell from that as well. So if they say, OK, they're using AWS EBS, then we know. So is this question redundant to what we already asked? I'm not sure. Well, I think this is the explicit deployment question that we haven't before addressed. Even if it isn't that way now, maybe we could reword both these questions to eliminate the overlap somehow. OK, so these two questions, should we just combine them as just as one, because there's some overlapping between those two? Make it simpler or? I don't know. I think it really the form of your storage is orthogonal to where you're located myself. In other words, if you define, if you use an outside vendor provided storage as a service that's one category, a dedicated appliance form is a second, and software defined storage is a third. And they exist regardless of where you're hosted on-prem or in a cloud. So I think two questions is warranted, but maybe not every permutation of those, right? Yeah, Steven, I agree with you in terms of the technical what would make a difference to us in looking at how these storage solutions would work. But I think the deployment, whether it's cloud or local or Kolo, gives us insight as to the closeness of the data. The expectation, in terms of performance, would be something that I would have an interest in. Yeah. And by the way, I also do like the question of going up a tier from block storage, where that what is it the fourth question down on 13, getting into asking about as a service forms of things like Cassandra, key value stores, whatever, as opposed to just getting block store and turning it into these services yourself. Yeah, we're definitely starting to see more of that. This one, OK. So for this question, Steven, do you think we could get the answer we want out of this just one Q13 and then maybe slide an option in here for co-location? Yeah, to be honest, my speculation is we can ask the on-prem public cloud Kolo, but I've sort of talked to enough people that I've come to the conclusion that with big enterprises, they're checking every one of those boxes right now. And I personally don't think I'm curious enough to know because I think I've already come to the conclusion, but we could put it there to confirm it. Oh, you're saying they're probably using everything anyway. Yeah. Oh, OK. You know, I run into a fairly decent sample of people getting around to conferences and whatever. And yeah, my observation is right now today, the big enterprises are going to be checking all those boxes. Maybe their roadmap is something different. But right now, they're pretty much everywhere. And I wouldn't feel I needed to ask. But if somebody else here wants to, I have no objection to it. OK. So we just to keep that one there then. Well, for the sake of brevity, I'm the main problem of having to share with that. Sorry, go ahead. So I think based on who this survey is going to, we're going to get different results, even from the people in the same company. But if you're on the outside of the house, you would probably know all these nitty-gritty details about the infrastructure. Whereas, I mean, whether it's Kolo or not, whether it's what type of block storage, whereas if the survey goes mostly to the developers, they probably would click object storage stuff or more cloud stuff. So I think one of the things we need to ask right at the beginning is what your role is. Well, it's for users. We actually talked about this last time. I think Quinton wanted to make this very clear in the beginning that this is only for the end users. So yeah, if you are a developer, then your role is for the user. Yeah, it's only end users. We're not talking about the vendors or developers. Even the developers can be a user of a storage as well. So if they want to. I'm sorry? Look, if it's like a... This is targeting your end users. If anyone can't click on it. Yeah, they could, but we are saying this is a targeting the end user. So if end user don't know the details, then they can answer, don't know, right? So... Yeah. The other thing, though, I think are the ones... End user is a very big term. Developers getting this. The objection of Kubernetes is to abstract out storage. And I could see that in some organizations a developer honestly wouldn't know what was behind the Kubernetes phrase. So that's fine, right? So then they just answer to no. As a... You don't have to be an ops person to deploy Rook or to deploy Ceph, you know? You... Whereas other stores, you know, you have to be an ops person. You have to get storage admin and IT admin. So based on who gets this survey, we will get different results. And I think one of the things we need to use as a filtering level is what your role is. Because I think end user is a very big term. Yeah, last time we have already got rid of that question. In the beginning of the survey, we do have a... Initially, the first version of the survey does have the categories, who you are. But then last time we... So basically you're using, right? So we changed this one last time. We got rid of the first question. So this is not... Yeah, but I think that one might be right that you can... Using it and then fall into a category of developer or ops. Right, right. Yeah, that's true. That's true, yeah. So basically this question, basically some people may not be able to answer it. Then that's fine, right? They can just say, don't know. We do have this one here, right? We can say, if you're a developer, don't answer, they cannot say that. That would be bad. But this is more targeting the end user. So... I think for question 13... Well, the thing is with... The boundary between developer and... End user definition is really vague and... Yeah, yeah, it is. ...page and... Yeah. I'm able to... I think, I think... I'll go for the earlier discussion. I'm sorry. At least the last time, what Quina was saying is kind of limiting those two people who are really, really the end user, not the admin or people like that. So... But it's hard. You can't say, okay, you can't fill in this out or something, right? So... Well, the thing is, and all the questions we're asking as far as consistency, availability, these are the stuff that an average user... You raise your quoting. These are the stuff that an average user wouldn't know about. This is the stuff that usually admins do diligence on and they figure out all the details. Yeah, that makes sense to me. That's kind of a... But I would have this... I would have... For some of the surveys I've seen in the past that we've here in our organization had success with is we would ask the demographic questions, those questions that would kind of identify who's taking the survey at the very end and make them optional. And that way it would at least give us the opportunity to gain that extra insight as to who was taking the survey. So... Okay, so maybe... Maybe I'll add that back at the end. We did have that kind of carry mode for earlier version, so maybe I will... Maybe I have like one or two questions at the end. What is your goal? We'd like to know more about who you are and what your goal is. Okay, so then we can finally when we get this, we can actually analyze that. But we're not going to add that to the mandatory section. There are already too many questions. Yeah, that's fine. Okay, all right, so I'll add this one. To go back to question 13, the feedback that I provided about COLO and data, Steven has convinced me that I think that... This is kind of redundant. I have enough value there to ask another question for it. So I think you can just delete that and just leave question 13 as it is. Okay, so, okay. All right, so I'm sorry. I think I was getting distracted. Should we remove this one? Yes, did my audio end? Is my audio working? Yeah, yeah, I was sorry. I was reading something else. And then I just wanted to make sure I got it. Yeah, okay. I think it's okay to just leave 13 as is. And Steven's right. We can infer from the answers to 13 whether locality is an issue to them or not. Okay, all right. So keep that. And then, yeah, so we still have this question. This is also kind of more advanced question. Data, how is data protected? So we don't see the student system though. A lot of users will know those details. But they can say I don't know. Those are pretty detailed questions. Yeah, we could probably tell by the answer to that question if they answer it, they're probably ops. Yeah, yeah, yeah. Yeah, the next one is saying what data services are provided. Yeah, this one is probably a little bit better than number 14. So, and 16, does your system require installation of a kernel mode? I think this module, this is a side. Bring this up last to me. He said he got the question a lot. So I'll have this one in this optional question. Could you read it to me? Oh, yeah, so question 16 says, does your storage system require installation of a kernel module? So this one was brought up by a side. Yeah. Yeah, that's a good point, yeah. Okay. Many, yeah, it's only an ops person would know the answer to that. But for those taking the survey, if they do answer that, that'll be good to know. Okay. No. But you probably want to leave an option. Don't know. Don't know. Oh, oh, don't know. Don't know, don't care. Don't know. Okay, don't know. Just FYI, I was just at a conference the past four days ending in Sunday and saw you talk on CSI. Oh, okay. To a big room and he asked this question. Oh, the kernel thing? Yeah. Oh, so what do people say? Essentially, if you're on software defined storage running in containers, it seems that the answer was uniformly, yes, that I don't think anybody heard of one that didn't need a kernel module. Oh, okay, okay. So it's probably, okay, so it's probably more more common. It really would be a question. If you really wanted coverage of the whole world, you could ask it to vendors of software defined storage, right? I mean, they'd know. And the users are in a position where they're not authoritative and they sort of maybe don't know or might even give wrong answers because they're confused. Yeah. Oh, the thing. Though I don't know, if the question had to go, I would actually drop this, but maybe we should give Saad an opportunity to defend it or see if he still wants it. Yeah, I think he just dropped off. He, last time he said this one is important, so I would just leave it there. I think that the right audience to ask this is actually vendors. Yeah, I think several of those questions are like that, you know, like the question 14, you know, 15 maybe more user probably know, but 14 I think it's pretty, only vendor knows, but the ratio coding as a user wouldn't be looking at that. Yeah, so I'll leave it there. I think some of those are from the white paper. Yeah, and then 17, it seems to me that if you're using an Orgis Raider, this has to be yes. Yeah, yeah. So, but some people, they still, they may not really know. I don't know if they, well, if they just use database, then they may not, then they don't use this, right? So if they use like... I suppose, yeah, if you're using it as a service or maybe you haven't even moved stateful apps of any kind to containers, in which case, you don't use storage at all. Right, but database is also considered a storage, right? So you, they're just doing like vitis, right? So then in that case, they actually don't use a plugin. So this plugin is more, you have to be using the block storage or something. Okay, and well, I understand 16 or 17, but if we get to a point of triage where we're, we've decided we're gonna purge questions, I put 16 and 17 on the cut list myself. Okay. Oh, I want to keep the number 17, the volume plugin. Okay. So, okay, and then of, yeah, this is a little bit more basically than we just want to know what type of a volume plugin you're using. Okay, so maybe I should read the question because I think, is it a jab? Well, he said he couldn't read it. So number 17 is, are you using a volume plugin to provision storage of containers? If answer is yes, then what type of volume plugin are you using? You know, Kubernetes inter-blogging, Docker volume driver interface plugin for like swelling CSI. Yeah, so this would be a question that I'm interested in, so I like to keep this as optional at least. And, oh, I jumped to the number. And the next question is what protocol is the block storage system you're using if that is applicable? You know, I just got to have a channel, I'll be there. And 20, question 20 is... You're gonna list those protocols. I think that was another thing that came out during this conference I was at scale. Yeah. But NVME over fabric was brought up spontaneously by a number of... Yeah, yeah, we can add that. As a future, not something they're using now. I suppose. Okay. So I guess if it's correct to leave it out, probably if it's what you are using, but... You'll... Oh, okay. You'll have... It looks like that's going to become big. Okay. NVME, you'll have or just NVME you'll have or do you want to be... Let me see how that's typically value. Because this is, the NVME you'll have is the one that we are actually in the plugin that I'm working with. We're actually kind of adding that. So... Yeah. Is it normally NVME-OF or at least that's how I recollect seeing it? I'm doing a web search now to see if I can... Yeah. So this is a particular one that we are adding. This is a particular protocol we are adding in our plugins. I know this one. But I think there are slightly different way of saying the plugin. It's a little different that I'm not expert in this myself. But I sometimes people just say NVME. Anyway, I would say added to the list. Okay. So let me... Actually, you know what? We actually have this one maybe in the wipe. I don't just do a quick search. Let's see how we... NVM. Yeah. We actually added this one here. So let's just use the term here. People will know. Okay. So I think here we just use the NVME. I think that's... What you're using it as a form of storage there as opposed to a connection type, right? No. This is basically the... Well, so NVME-OF would be a protocol. NVME, so this is a transport used for the... So used for the storage, right? So you can use that with SSD. I think NVME part of good. I will search it a bit more. But... I think NVME-OF makes sense. And I don't think it makes a difference whether we hyphen it or not. I've seen it hyphen to... Oh, yeah. I think the way I'm writing this is correct. I'm just saying, I don't know if... Do we... I think there's maybe an older form just to NVME and... But yeah, anyway, this refers to connecting to it in a remote box over Ethernet. Even though technically there are ways to do this over Fiber Channel and other transport, but... Yeah, okay. I think it's probably okay. I just add this one. If people want to write something different, they can say, you know, it can say other or something. If they say, think this one is more too specific. But this is definitely a protocol that they... I just don't know if it was a broader way of stating it. So that includes this one. And there was another one. So I'll just leave this one here. But if you guys think we should really... I'm in favor of putting it on the list. Okay, all right. Okay, so the next one is the file. Share the file systems. And then this one, just in NFS, SIFS. And then that's almost it. And are there anything else you want us to know? So that's basically a very general question. If they want to just put on any more thoughts they have. So that's... So a little bit over 20 questions in total. Probably 22 questions. Or maybe 21, 22 questions in total, yeah. But if they just want to do the first trial, then maybe they can get it done in 10 minutes. Yeah. For 21, I think that's a good opportunity to have people mention any other technical challenges they're facing as well. May want to suggest that. Are there any other... Anything else you'd like us to know or any technical challenges you'd like to share? Maybe try and elicit a response from folks who are having technical challenges. Because that's kind of what we're looking to address is what are the technical challenges, right? Yeah. Yeah, so if they couldn't get a covered then we have that ranking question and they can elaborate here. Is that one? Yeah. That one is just like you kind of rank it. But this here you can write down whatever you want if you like to. Write. Yeah, so. And then the last one, I just need to... I think there's more categories that I have in the first version of this and I will go back and check. There's categories of the rows. So I'll add back here. Yeah, we'll have to figure out what's the right grouping there, because... Let me see. If everybody just puts in user, I still would be at a loss as to what their role is. So they can be... Yeah, so this is... We'll make this a multi-select because people can probably in multiple categories, right? Okay. Yeah, they can be a... Yeah, there is a... I need to... Let me save this first just to make sure that not... Yeah, I just want to make sure that I don't screw this up. So I want to just go back to the history and there's an earlier version that actually has some categories. How do I go to earlier version? Do you guys know? How do I go? All change, every change. Okay, let's see. Last edit. Maybe this... Okay, this will tell me because there's an earlier version that I know have a... It's probably a little earlier. Okay, this one probably have... Yeah, okay, so I just copy this and then go back. Wow, that's useful. I need to do that more often. Yeah, so, okay. So now hopefully I'm in the... Yeah, so now I'm in the latest, I think. Let's go check. Yeah, it says Alex, right? Okay, so let's see. All right, cool. So we have... So obviously you're a contributor, a vendor, distributor, cloud provider, cluster operator, cloud native storage user. So your app developer, deployer, something like that. Maybe we should, okay, move this one to the top. Since this is targeting the user, so we just make this the first choice, highlighting that, but then they can select other choices. Hold on, that looks a little funny compared to the others too. Which one, I'm sorry? Oh, the fact that cloud native storage user is in bold phase. Oh, I just even want to highlight that one. So that's our target groupers. So if you would just see that one, and saying that we are focusing on the user. And so a CCNCF project cluster operator, I don't know if somebody would self-select as a project cluster operator. If they were just a user that, you know, we're sysadmin managing their own cluster. Right, so probably, so some people probably would select both. So the user, I think it's more like an end user. And then this one maybe. Why can't that just say cluster operator? I'm not sure why. Which one? Which one? Here? Yeah, project cluster operator. Uh-oh, just a new one. Oh, okay. Yeah. Yeah. Yeah. You mean get rid of CCNCF? Is that what you say? Yeah. Okay. Cause I don't even know what that means. If you put the CCNCF part of it. Oh, I think a part of it just means that we are, okay. Just Kubernetes or those projects doesn't mean anything. Yeah. We already asked whether we already opened it up to using Meizos or cloud cloud. That's true. Yeah. So I think it should just say cluster and. Okay. So what about, uh, we have, okay. So what about the others? Yeah. So this one would be like a destroyer. So, uh, you know, it could be maybe open shift or if you are building that or something. That sounds like. You know, that current CNCF, um, I chart of the, What is it? The landscape now has 600 plus projects. So. Yeah. It's kind of too much. You don't know. I don't know if somebody is affiliated with one of those projects or, um, I think so this. So those categories originally come from, uh, some survey. Uh, the, I think that the maintenance team or the, what a stable maintenance team. There was some team. They, they have some survey. Uh, so this is originally template for Quintan. So this is the first question. So some of those, uh, uh, roles are from that. Um, so maybe that's why in a small targeting or CNCF project. Yeah. I sort of, um, if I were answering this, I wouldn't know what it meant. And if I were trying to interpret results, I wouldn't know what it meant. So maybe. I want to remove this. I can remove this one. Yeah. Okay. Distributed because I'm, yeah, I don't really, I just saw that my size. Okay. I'll just have that. Uh, so the rest of them should be straightforward. I guess. Yeah. I think the others are fine. I would say end user. I would say end user. End user. Uh, Yeah. Yeah. And then develop her contributor. Uh, you know, vendor. Club provider. Right. Okay. So maybe that's better. Yes. And that might be a bit weird to the distributor thing. Maybe because they're more, that group is more focusing on, you know, what is the stable maintenance release or something? Maybe that's, you know, they, they are more, uh, they want to know who are the distributor, right? Because they, they would be, um, care more about the, the maintenance release or something. I don't know. Okay. So, okay. So does this look good now? Yes. Okay. All right. Okay. So, um, Uh, okay. So I will just update this, uh, based on what we have discussed. Um, anything else? Yeah. Not for me. Okay. Uh, anyone else? No, I'm excited to, to see some responses. Okay. So probably I will need to, uh, I just need to put them in a survey. I will ask. I'll check with Alex and Quinn and see if they, uh, what they want to do before I, uh, move this to a survey monkey. I don't know if, uh, I think we're supposed to, uh, uh, to talk to the end user group as well and see, but I don't know if they want to just look at this at this way or do they want to see it directly in the survey. So I'll check with them. Um, but at least now I think, uh, I think this is in, uh, a good shape now, a better shape than in the beginning. Hopefully they can be, hopefully there are more people who want to, uh, feel out the survey. Yeah. Get some results out of this. And how are we distributing? Oh, so it's going to be, uh, converted to a survey monkey. Okay. Yeah. And what list are we sending it to? Or it's going to be the end user. So I need to find that, find out from Alex. There's a, there's a Cheryl is running that group. So we need to work with her on how to distribute this. I think I don't know. It's just mainly listen what I don't know. Yeah. And is this, uh, uh, something that we could then share with our colleagues that we are managing clusters or end users of clusters. Uh, yeah, of course. Yes. I think so. Yeah. Thank you. I think that would be helpful if, yeah, that would be great. Yeah. Yeah. I think typically for some of their other surveys, they all use an attendee list from conferences and things like that. But I think we, but I think we want to do this on tweets and things too. Yeah. Yeah. I think that we, yeah, that would be a good, I just see that when you register for the conference, they actually ask you certain questions. It's already kind of a survey feeling on the survey. But, but I think that would be a little too late for us. Because, right, because we need to, uh, get the survey out before the conference because we want to actually be, want to be talking about some of the survey results at the conference. So, um, we actually want to do this one before that. So we'll see what the Cheryl says is she's running a group. So, uh, probably will need to send this to their, they have probably have a meeting is sending out and then, and then ask them to feel out, I think. Okay. Yeah. Yeah. Yeah. In any case, I think once this one is finalized, this will be shared on the, on our mailing list as well, right? The storage wouldn't go with willingness as well. So, yeah. All right. Um, if we don't have other things to discuss, we will have a 12 minutes back. Thank you very much for your work on this. It's been several weeks you've been working on this. So, uh, definitely we appreciate it. Thank you for your feedback. Thanks. All right. Talk to you later. Yeah. Bye. Bye.