 There we go. So, um, so, yeah, I want to then do the November 25th sick meeting, uh, two more for one year anniversary. Is it? Yes, this is a one year. Oh, nice. We started. Yeah, this is the 26th meeting. Yeah. I think we might have missed the meeting, but I mean, it was literally the first meeting was. Well, it should be very basic. November 26th, last year, so it's the eve of our birthday. Yeah, let's celebrate tomorrow. Okay, so we all know each other, nobody's not today. So let's go ahead with it. Yeah, it's updates. So I wanted to talk to anything to him as well about I progressing iconography right now. We are focused on table studio stuff. We are not able to support any at the moment we expected to do in the future. The iconography work, but team I see you have already started. So the situation I, there was a situation that we, we discovered that there were some libraries being used so that something needed to be addressed, which is, they will handle its own icons. They will handle through an external library called icon set. And they can see package. So, and there's a PR by team to integrate the library into anything so that we can start migrating the icons. How do you, how do you think this is going in. So it works. It's just a test failure. The feedback on the PR was good. Put some stuff on there was all fine, but we just need to look at and see why the test is failing. Yeah, just bring me I want to pop this one as well. Okay, so. Um, so that was just a generic update. We are thinking about iconography team is basically there in it and trying to support him as best as they can. And I wanted to talk about what we talked before about the iconography work order, something that I don't have screenshots at hand to share with you right now. I will try to dig some up. Well, we mentioned we spoke when we first introduced the iconography work that there were several icons, several items we wanted to to restart right and we categorize it in several different categories. There were the. So yeah, let me just go ahead. So we talked about first about the build status icons, the famous balls, which I don't see any. Yeah, these famous balls, we, in fact, Joe provided some designs to improve them. The idea of the migration was news idea was this one. I already bus team these assets. I think in the first batch of iconography changes we can do instead of changing the whole UI we can focus on these items. And also in the weather icons. I've been thinking a lot about this. And I think it's my personal past hate in the Jenkins your eyes. Yeah, because it's it's it's going to be tricky to make them look to make these colors good in the SVG but also make them theme aware. I don't know yet how to do how to better approach this. But what I think is, I've been thinking a lot about what it would mean to change iconography in Jenkins. And this is the conclusion that I'm going to share my opinion on this I expect feedback, feedback that maybe for this will be a cycle, we should just focus on this, on this is the status indicators, and the weather icons, because from what team, at least for my experiments, and I think from what I see in teams branch, and what they say in this branch. So the generic items, for example, for the, for this side side bar entries for the manage Jenkins for everything. If you change the default items to material, there are many plugins that will not change. And there will be a weird period where some icons are modern using material sharpies with the icons, and some are just not. That's something we could maybe not do without for these ideas for this, because this is a release that will. Let's face it, we'll have weird UI stuff regarding the forums. So, and so, yeah, my proposal would be for these ideas cycle just focus on the status indicator icons, and where the icons, and if you get those right then for the next one. I don't think it looks pretty bad. It's not, it's not, it's not that obvious that they've changed. They just look sharper. So if you look at the new icons, and, and say, corporate requests that I made, I don't think I'm maybe maybe we leave the slash manage icons alone as those ones are harder to deal with in the first iteration, but at least the ones on like the homepage. The icon's changed, but it shouldn't really be jarring. I think this is something. Yeah. The main concern is that there are many plugin defined icons, right? Yeah, at least if you do the build on the weather, I mean they're kind of localized and one part of the page is not like in your site navigation. There's other icons where the weather icon comes, notably the folder icon. There's also GitHub or branch source kind of ones as well that look a little bit different on the display next to the weather. So if we're localizing it to kind of like the weather and build, as in just looking at the main panel, I think there's other plugins that still should be fixed for the main panel to give you a seamless experience there, regardless of anything happening in the sidebar. And I have to update plugins in all plugins are going to have to update anyway, regardless of how's it, but I still don't really see the reason to restrict us to I mean where there is a good one to focus on. But I don't really see any harm and update in the side panel as well. I don't know. I would still focus on the list of the syndicators first, because it's, it's nice. It's a, it's a focus located set of icons. And I just, you know, together. You know what, know what else. Well, they are using the when building the updating plugins on the plugin. Also use at least the Legion page. Sorry, use on the Legion page as well. Yeah. Yeah. Well, I mean those are specific subject to fight on survival to start with those. And then build up from that from inside out, right. That's what I would suggest. And then we evaluate each and every after every PR. What's the jar, the jarring effect. Yeah, I think for me for me it's really weird sometimes in. Yeah. Yeah, I still think this rating is better than restricting. Starting with the whole request that just changes it for one icon and then iterating out and one or two icons and then iterating adding it to more. Some of them are in cool or some of them are plugins. And a lot of them is just choosing the icon as well. Yeah. And I was choosing the icon regarding choosing the icon. So it was, I did ask all about this. But was there are many icons that are, for example, you can just change these people icon to a bunch of people right there is a terminal icon, there may be an icon for this. There are some icons that they don't necessarily have a matching material, for example, or in any popular icon suit some items that definitely need to be custom. For example, I don't have an example here but there are like 100 variations of a notebook icon, a notebook with something right. There are many, many different variations of those. So I did ask Oleg about it. He told me we need to follow up on this, but maybe we're going to launch and put a blog post and request for contributors to contribute basically icons to get some iconography contributions from the community. And maybe launch us, I don't know, swag campaign or something. So I think there is going to be maybe an opportunity in the future. So this is something I want to follow up. But I think that would be a good source of upgrades for for customers that necessarily have a match in a library. What do you think about this? Yeah, it makes sense. Anyone thought what's happening with the animation on the currently building? I mean, the Google Doc shows it as flat and not animated. I don't know if you're planning on there will be animation or not. But but the experience I have of using any animated SVGs on Windows is all browsers suck and they choose CPU as soon as you start using animation via JavaScript or CSS. Yeah, I think it depends because you can create a gift file, which is not ideal. Especially on the last screen slide. It's not ideal. There are several ways to animate it. I would go and animate through JavaScript. There are three ways you can animate it that I that comes to my mind, which isn't the SVG code itself. You could add a rotation property on the SVG itself. Through CSS, I mean, a rotation of property and rotation animation in CSS. I think those two should be the same in performance. But maybe in having it in the SVG line, it's a bit more, it's a bit faster. I do not know. I'm not sure about that one. I need to do research. And then there's JavaScript animation, which I would rather use a key rather than use a JavaScript animation. But in nowadays, I mean keyframe animation for rotation for just one element shouldn't be too hard and to impact too much. It's not one element though, it's when you've got a page of builds and you've got a lot of builds running. Each element adds up and we had this in Jenkins X in a proprietary UI that they were doing and I could spin my browser at 100% on one core just to do animations. And it was kind of like, it's just horrendous. Yeah, let me write down, I need to follow up on that one, okay? Yeah. Yeah. I'm going to write a note in the minutes that you follow up on that giant fix. Yeah, put something under action item, action item world. Okay, that's it. Yeah, we need a solution for that. Okay, that said, we can always follow up to a give. Right. So, what's going to make maybe or don't animate. Well, not animating stuff in progress. It's difficult to convey stuff without motion to convey progress without motion. Let's just do the research. I mean, I mean, obviously, you know, do it tested on Windows. Yeah, let's, let's get up here going on and throw it out to the wolves. Let's see what happens. Okay, so in the same one want to know more about economics, by the way, up to now I've always been talking about material icons and everything. I'm using SVG sprites. I'm not 100% married married to it. If a better solution comes up. Because I'm sprites was the best thing I could do at the time. It was the most pragmatic one for me to start working with. We can replace those uses in your score. So if anybody has a better idea. Just a stupid question. I mean, do we just use material icons because that means that we get the icons for free, or are the other benefits of material icons except for that they have a big library for us to pick. They do have a big library. I'm getting bigger, but the sprites, SVG sprites are not well supported. There are community fork to provide the sprites. These are the sprites and they, because Google is going back on for the producing icon font, but I know some people like to use bootstrap but I think material is more. Material uses an icon font as well. So I mean over just getting, you know, I'll sign it to make a bunch of images. Yeah, basically, I mean, what we need 100% in my opinion what we need is a open source icon library. We cannot go with customers. I'm not married to material. If material becomes a problem, I'm more than happy to revisit it. Yeah, that's why also we should give the changes containing the Jenkins core. I'm not this college. They are the plans to bring their own icons. So any thoughts on this. Yeah. Yeah, I think they have much more icons. So this is the reason that I'm using font as an and they are more used in web applications and material. They have, yeah, a lot of new ideas, for instance, to in the two color icons in the next release will come. So I think they are more progressive than material material is something. Yeah, it started from Google, but it is. Yeah, I don't like material design. So maybe that's, I don't like them so much. So I'm using font as some, but actually the most of the main problem we will have is that we need custom icons. And it's really hard to design some if you are a programmer like I am so they will look a little bit strange when I'm designing an icon. So I use font as some because they have the most icons. Sorry, how do you call this. I'm just the name of this thing. Font awesome or as my own it's not awesome. Yeah, I know what you mean. So maybe we'll do that. I have a point to maybe revisit also using shopping material with font awesome. Whatever it's more comfortable. I, what I like about material is that icons are common, especially with 100. So I think that's something we can revisit. And I think this, this brings back to my point about maybe we need to get a little more weighty or well, whenever the people in the new parties isn't asking for this task in charge of communication and everything to try to maybe get some community involvement, community contributions regarding design and art. So maybe that's something that's already my opinion. Okay, thank you. Thank you. I, I sort of agree with you. 90% with you personally. Okay, shall we move on. Yeah. So, James. No, yeah, James tables to do you have had your concerns regarding the tables to do some of the impact it has and basically breaking many stuff. So, yeah. So internally with there's been quite a few plugins sort of broken. And I think there's been some open source ones that are broken that you've found and fixed what you know about. I'm concerned that this is going to break stuff that people don't know about because people write custom plugins and they'll upgrade and things will just be broken but not from a UI layout point of view but from a form submission point of view is what I'm concerned which means if they kind of go to change something, hit save, then they could have lost some configuration in an unrelated part that they haven't changed because it's missing from some form submission. That doesn't seem very backwards compatible. So, I'm just wondering why we're not trying to say this is Jenkins three dot X, at least signifies that there can be some breaking changes and people need to do something and it might not be as trivial an upgrade as as expected. We did a Jenkins two dot X that all that had was a startup wizard and didn't break anything. I'm just thinking with with the table to give work and also the work with a security as well that has been going on elsewhere. I was just wondering what people's feelings were on kind of bumping to three dot zero one or whatever. I guess by failing on that one is a little bit you know the journalist phone us up you know, you've released Jenkins three. What are the big new features and it's like well we've transitioned from table steps for, you know, it's a little bit you know, I understand that it will maybe make it clear that a lot of plugins could be broken and it'll be really important to upgrade but I just feel, you know, from the outside point of view some is going to install Jenkins free you know thinking finally some new features and it's going to look just the same as it did before. So, Chuck some like as well, and then it looks pretty. I need to talk about this with all of you tell me the best thing would be to get a meaning is that going on. But I want to put words in his mouth but in my conversation with him I understood that he he thought certain concerns that that's something we should have supported before. Which doesn't deduct your point games. Definitely. Yeah. I don't know how to feel about this I think you insist lacking a major release policy. For example, for example, having a calendar for a major every year and a half. So formal deprecation notices from one major to another. You can remove a staff of it, but it was tough every year and a half or every three years. So I think that's a policy that is lacking and should be in place, maybe that's something we can raise. But that's a separate thing for the industry. So, if we want to, if we want to follow up this, maybe raises this question with the community. Maybe get a meeting is going on how, how is the discussion for this. How does the discussion for this happen. Maybe the dev mailing list would probably be a place to start. Who would even make this decision. I mean, it's not obviously it's not your exit that's going to decide that we're going to go to. Community. Eventually it would probably be the governance board that would say, yeah, or no, but it's up to anyone to propose it. Yeah. And sit feedback. I think it's worthwhile starting discussion. I mean, maybe I'm just a little bit old school, but I mean, I kind of, you know, remember back in the 90s or the early zeros you go from version two to three and you'd, you know, install it. You're looking around to see what all the new features were. Yeah, but now you go from ground. Yeah, that's different. I mean, then us going from, I mean, we went to version two, five or more years ago, five years ago, I think now. I mean, of course, lots happened between, you know, two dot zero and three dot zero has been huge change, but there's no big new feature. I think I think it's worth it. At least raising the discussion, because even if it's not for now, which I see that there are some compatibility concerns that should be addressed maybe at least the communication maybe should be improved. I mean, it's has also mentioned on the has also raised an issue in the community board and on the issue tracker. I think that's right. So, and I think at least raising the discussion, we if at least set the set date for a genkis fit free or a discussion for an official major release schedule comes. That would be great, in my opinion. And so, again, if you don't mind, maybe creating the discussion for the, for the, for this. And maybe this can be a hot ball. How do you say hope for the new work, as long as they, it's in state. Okay, thank you. Yeah, but we didn't ask the question. How did you feel about the whole V3 stuff. So, because I don't think you feel I'm not sure. There is that's made a catchable that. Sorry. So what did you say before. So, you know, how do you feel about these warren being a V3 version bump. Not sure. Like the community plugins. That are there. It's quite a few, it's quite a few of them, but they're all low installation ones. It's the more obscure ones. And we're, but the problem is we're seeing completely broken UI pages, which not sure if it really bump into three and not sure if it really helps that it's really just need to fix the plugins. Yeah, it's it in itself isn't going to fix anything. But certainly I think when people upgrade to a new LTS or take a new weekly, well probably not the people that take a new weekly because they are the people that like living dangerously. But the people that, you know, move to a new LTS might skim through the release notes but other than that it's it's just to upgrade the war have to get upgrade whatever. And, and move along they generally might not expect to look through all of the release notes or do anything else so so it's just the fact that the three dot zero. And that's all that this has got a, you know, if you go via the same where which Jenkins does then this has got breaking change in it. So maybe I want to pay a bit more attention than just trying to live once. Jenkins does voice in the none of the plugins ever. Jenkins code itself kind of does, apart from this recent change and the other one because you can take a plug in that was written 1011 years ago and run it in a Jenkins today and it'll still run. Apart from the Jenkins to which was just marketing bump. That's why all the old deprecated code is still in Jenkins course because we can't remove it because it's like, we're compatible with something that was written ages ago. Oh, it's slowly. And none of that's done major version bumps. Anyway, I'll propose that we can. Okay, I think I think it's going to be a fun trip. Okay, great. So, next item, I want to do a bit of a bit of back triaging for them, basically, team and I will special team are having. There are many reports of coming from the most of the stuff. And users just say my configuration page is broken. Right. So basically what Jen said, right. But it's not really that helpful. We need a list of plugins. We need something to see how is broken. So, but then there are some those parts can really be a soft plugin side. I myself, I move focus on fixing the stuff within Jenkins itself. Right. There are some bugs that are present in Jenkins, instead of violence. For example, there were some, some of them that calls with regulations introducing 244 for context genius 244. Introduce the tables to leaves and JavaScript 20 versions in advance basically. And so it's present in two of these cycles. There were a few bugs I found and I tried to fix. So I include them here for the sake of showing them because they are back in Jenkins core and actual regressions in the, in the current BS. And there's another one that back raised, which is a drag and drop. It's not working good. So dragging drag and drop, especially from dragging an item from up down or something is not working well. So I'm going to be, I will try to look into this drag and drop it doesn't really work well with really big items. So I think it works on all the smaller ones. So I've tried to do things and I haven't seen any issues. Yeah, to be honest, I just want to maybe remove the stop emigrate the terrorists and because it's happening, especially in two popular way with you. It's called a repeatable. Part of me just wants to replace the drug and drug implementation with you. So we're completely in favor of that. It's an ancient prototype based drag drop. Yeah, it's based on your UI and from what I've seen the UI code, it's the class is not instantiated in any, at least in the European source plugin. And we can, we can always leave that all the code behind, right. There's nothing to avoid. That's nothing that keeps us from from from from leaving the code for. Whoever wants to move it if we can, I think. We don't need to know this. If we can't find it being used, then we just remove it and we can always add it back later. Yeah. So I personally would be in favor of. Just changing hetero list and repeatable items. Hopefully tables to do this way from the tables to this code search and everything we found we have found uses of custom custom implementations on hetero list. That can be at least we know they are there. Right. So, I need to be looking into this first of all I will try to fix things. Maybe I would even try to add some small buttons to bring up and down the elements so that you don't necessarily need to drag those. You can, you can just click. Yeah, so I can drop was not a common user interaction. I don't remember seeing it on any form on anything modern. And to be honest, I think some nice arrows would do the trick. But personally. Okay. So that's, these are the two backs I want to mention, and there are other regressions is there any regression you would like to. Those are the ones they wanted to, to, to bring to attention because I think they are once in Jenkins in actual Jenkins itself. I'm not in plugins. This is the current dashboard for the regressions. I need to place it in the chat the enemy for context. Is there any other outstanding back team that you can think we should focus on or that's especially tricky or can do without with stuff in Jenkins itself. I mean, 90% of them are really easy. They're under, they're usually under an hour's work. Just to fix them up. They've got tables. They're using the tables for their form layouts. It's a IRC bot and plot, I think they are what they kind of need some redesign to work better. So the IRC and plot and more work than the other ones. Most of them are trivial upgrades. We tend been doing things like documentation migration and parent palm upgrades and incrementals at the same time. Just, these are all, most of these are plugins that are quite old. Many of them are on parent palm two or one even. So some of them need spot bugs fixes and whatnot. But it's all quite straightforward. What I wanted to call out is the TFS plugin, which was suspended about two weeks ago for licensing issues. TFS team foundation services. That's Jason object Hudson triggers SCM trigger. So that plugin was suspended by Daniel because of a licensing issue. So it can't be downloaded anymore, but there's still some people using it. At least mobilization who want it back. Okay. Are you coming through? Because I'm seeing you because you're the one who's dealing more with with the public side of decision. Do you find it difficult to get maintainers involved or are you finding that you need to produce yourself most of the time? I haven't seen any maintainers fixing them. I've seen a few of them have been fixed by the community. So, yeah, a few of them have been random people coming in and doing it. That's had it. I've never seen them before. I haven't seen any maintainers fixing any of them. At least the ones we've got to this point, they might have fixed them earlier. A lot of these plugins are un-maintained. So I would say the majority are un-maintained. They're usually the older plugins. Things like release plugins. There were some plugins that were really with low use. Run deck. I adopted some of the ones where people were, the ones that were marked for adoption because they're easy to take over. And one of the ones where lots of people were complaining about. But quite a few of them have just got requests sitting there. I think the only cloud-based one is Active Directory. I think cloud-based maintains that one. We will look into that one. Did you manage to get an answer from the Azure DevOps team in China? Yeah, he got back to me and said that he couldn't do it last week. I'll chase him up later this week, I guess. Because most of the Azure plugins are broken. One thing to be aware about with Azure is that they basically shade. They've got a common library which they share between all the Azure plugins. And they basically shade it. And that's the one that's got the tags in. So even if most of the issues are in the core library. But even if you upgrade the plugin which the library is in, it's called Azure Commons Core. You'd expect upgrading Azure Commons to fix the issue. But it doesn't. You have to upgrade it in the plugin as well. It's quite annoying. They use mask classes to hide things from the class loader. And it's quite painful. So yeah, there's your ones of it special. And they've got really painful dependency management. Well, what's called the answer? Yeah, the other, you said, the Artifactory plugin, which was the only plugin installed by more than 1,000 or 2,000 installations. That was fixed today. So I think that should fix quite a few people's instances. The other thing is that if you go to the top issue, one with 36 watches. Yeah, this one. So if you go to the comments. There's like 70 odd comments and there's lots of people who haven't managed to identify which plugin is causing the issue. They've mostly posted their plugins. And it needs triaging. I mean trying to get them to do it, but can be hard. I will give it tomorrow. So if anyone wants to spend a bit of time triaging some of these, it will be quite easy. I will give you a hand. I will try it tomorrow morning. I'll pause triaging for now and just start fixing some of the ones. Yeah, I will try to help you. Let's see what I can do. Okay, so yeah, thank you for this topic. We are 17 minutes early. So we have James here and I would like to introduce something, an issue. If you don't mind, I'll add an extra agenda topic. So, a blogging manager improved movement effect. Apologies for adding the topic right at this point in the last minute, but something I came up with. So, for context, for those who don't know team has created a really nice PR to greatly optimize the search and to do the plugin search in a way that will page server side and generate the listing JavaScript. And given that James has done lots of work, do you have a screenshot or something? It needs a video really to show it. Yeah, well, it's okay. The point I want to make is, not that we have James here, maybe James, I know you know a lot about the ATH tasks, especially how they are set up and everything. So, did you think maybe, so basically what this does is changes the search, when installing plugins, something that I know many, I think many ATH tests to, did you think this is... I don't think it will break it. I think it will just get a lot faster. A lot of times spent in ATH on waiting for this page. Definitely. Not anymore though, because it used to spend ages because when you went to the available tab, it had to load up all of the plugins and render it. And now it doesn't. So it comes up almost instantly and you say, I want foobar plugin, you type in foobar plugin, it searches Jenkins and you get the list back of one. And it's so it's much, much quicker than it used to be. Do you want to back version of the update center or? Just say that again. How does it give the shortlist? It makes query against Jenkins. So Jenkins has got the full update center list and nothing is displayed in the available tab until you actually type something in and the point at which you type something in, it goes, I'm pretty sure it goes and asks Jenkins. No, it doesn't. I'll behave. So Daniel changed it to not show everything on the page. So Daniel changed it to not basically it was a humongous page because it had over like 1800 plugins or so on the page. But Daniel's change was basically to display none on all of them. He still loads it from the server puts it in the page. You got to wait for this wait for the server to render the page and send it to the client. And it's still slow. So I showed it. So it normally takes about two to four seconds and sometimes when it's slow, I had, I've had to take up to like 17 seconds from a full page, like fresh cash. And this takes it to about at least five, six times faster. Okay. Yeah, it might have been quicker than it was before because your browser didn't have to render it anymore. But this might change it to just load the top 50 plugins. And then the rest of it is searching server side. And the other major change is that it does is if you do an exact match, it will bring it up to the top. So things like get plugin and blue ocean plugin work a lot nicer. Okay. Most of these reviews is no one seems to want to approve it. Yeah, why is that? Oh, I don't think. Do you ever, I haven't heard it, but I do really requested it, but are there any outstanding concerns? I don't think so. There was only one concern from Gavin that sorting only sorts the client side table. And I didn't really think it was needed to change it to sort server side. That can be a small follow up. Not necessarily a blocker. No, it wasn't a blocker. It was just a, let's say what people think and we can enhance it. Change it. Okay. Yeah, something I will, I will try to look at. He's able to try to run some eight age. We saw me pH suit with this. See, it doesn't matter and see how it, how it actually looks and post maybe a few videos compared it side by side some tests. And for my, when I've last run a th tests, it just goes through the UI and loads the page and uses the search and clicks. Yeah, if it clicks, and I haven't got around to fixing that page yet, it'll be completely and utterly flaky because Selenium and clicking is awful. And if there's JavaScript involved as well. So you need to kind of then wait to know when you're done and ready. Yeah, that could be the only issue. My concern here would be that the reason I wanted to ask you some of the stuff is that they can, I don't know if there's an issue that right now, what the job does is it filters clients. Right. So maybe I don't know if it's within the same within the same job to go and look before Selenium search. Right now, if the age needs to go to a server, maybe the Selenium Comprobation, trying to click it will happen first before Selenium. When you click in Selenium, there is no guarantee that it waits for absolutely anything whatsoever to happen. Not even a page reloads, not even anything, unless you're implicitly waiting on every single action that you do that we don't do in the acceptance test harness. You have to know that you are going to go off that page and so wait for an element on that page to become stale and the new page to load. Or you have to kind of wait until the new element that you expect to appear and you have to know what it is that you're expecting actually appears. And if it doesn't appear then after 30 seconds or however long your test will fail because you've waited for something that you now expect to appear and hasn't. In your framework or just, they'll have their issues though. Yeah. Yeah. Never use the UI one doesn't have some sort of weird issues. Yeah. I'm slowly getting through trying to fix a load of timing and kind of race conditions in the age, but it's sometimes it seems like a losing battle. Okay, I have, I have an issue, something I want to do, which is changing the behavior of the unchanged. That needs a bit of redesign of the sticky form buttons. Start using making them sticky or less even less sticky but using CSS instead of JavaScript because it costs a lot of test failures. Because Serene will try to pick something, but the buttons having been moved using JavaScript. So the form elements are hidden behind the buttons. In a related note, we had the same with the breadcrumbs earlier in the week. So, so the breadcrumbs had wrapped, which meant that and the tabs when you were in the config for a freestyle job had wrapped, and the button that we were trying to click was somewhere underneath the breadcrumb bar, because the scroll had gone. Oh, it should be about here. I recall, I recall having an issue with it. The buttons were overlaying something with for exactly one pixel. And Selenium only did a confirmation of greater than or lesser than it didn't handle the greater than equal equal. It was fun. 10 hours. Okay. So what I will try to do tomorrow is I will, I will try to run some to run some ATH using this. It was using this and see how it feels and see to verify that doesn't break anything. Because, yeah, my concern was leaving everything going, having a separate side of the country, maybe would mess something with Selenium, but I try to come back tomorrow with some actual data. I think hasn't the ATH been turned on for core PRs again now, but only on the core ATHs. I think it's only core. It runs smoke tests, which is like two tests. It's the view test and something else. I doubt it installs any plugins. Right. Yeah, that's not core behavior. Yeah, it's getting fast loading basically. Yeah. Okay. Okay, so. Okay, so that's it. Apologies for adding this last extra point. I thought, well, we have some in the next 10 minutes. So is there anything else anybody wants to share? Okay, then. So thank you very much for coming. And tomorrow let's wish the, let's all wish tomorrow, happy birthday to you. Yeah. Okay. Thank you everybody. Bye bye. Yeah.