 I'm gonna place it. So on the agenda for me is just looking at some issues. Other than anything else you wanted to talk about for Iyaka and Kushal in particular. Demo stuff or just quick updates. If not, then looking at 11.3, I'll share my screen, we have these things here. So let's talk through them really quickly. I think all of them have MRs, is that correct? Yes, so the first item of the list is something that we found in the afternoon. Actually, Mark found it on staging as well as happening on local. I mentioned a GIF within that issue, mentioning what is happening there. It was a small bug fix. I have the MR open and it is currently assigned to fill for review. So it will mostly be reviewed either today or tomorrow and it will get fixed. So the problem was that if there is any fixed date set to a particular epic and if we switch to date type as from milestones and then we switch it back again from milestones to fix, then it would replicate the date whatever the milestone had into the epic one. Okay, so you marked as 11.3, that's great. And then let's talk about this one last. These two have open MRs, right? The last two, is that correct? And then so we're just waiting for the review. Do we expect it to be merged by Friday? Which is the seventh, I think. Any problems with these two? No, those two will be merged by Friday if really something. I expected it doesn't come, but it shouldn't. Okay, thanks Sarah. So reviewers are looking at it right now and then you folks are addressing comments. And then I know this one, there's a lot back and forth for the bug. Yeah, because the one issue is fixed or this is just one thing I have to do to provide to the maintainer before I merge it, before he merges it. But the original issue, there is discussion between Jan and Alessio and they are trying to figure out what is actually a good way to fix it. Okay, so should I bug Sean about it, Yarka? No, I'll talk to him about it today and we'll solve it somehow. Okay, so you're still managing it. I understand it might not make it for 11.3. Yeah, because there are two possible fixes. One is on Vargos site and second is on GitLab site. And maybe we want to do both, maybe we want to do only one of them. So we'll see. But at least if we decide to solve it on the GitLab site then I'm pretty sure we can make it up to Friday. Vargos, I'm not sure how the process is there, but I'm pretty sure. Again, it shouldn't be that difficult. Okay, so it looks like everything will be finished here except for this one, which might bleed into 11.4. So let's take that into account as we talk about 11.4. So I like to this in the agenda. So basically I tried to classify into two sets of things like the do-22nd. And so if you look at the do-22nd, I basically want to see if we can finish all the open and close epic things and focus on that. And then maybe squeeze one extra one in, if not, then that's fine. But let's just look at this list. So let me go through this really quickly. So closing and epic. So you need to do it on the back end. Close and epic. So let me, sorry, let me step back. So when I say close and epic, I mean exact identical functionality as, you know, open, close for issues, right? So pretty straightforward. And so when you close and epic on the back end and API, that should be hopefully pretty straightforward. So closing and whichever one we do first for these two, closing via buttons or closing via quick action, we have to also do the UI design or the, this badge, we have to at least have this badge, right? So, so that's the front end thing I would see. And then, but so, so that's necessary. That's only state I can see from the front end perspective, but obviously you have to have the functionality for button version and the quick action version. And then from the list view version. So you, we would need to have these three tabs essentially. And then there's a discussion. I don't know if Pedro responded, but I said, like let's not even worry about the counts for now and just have these three tabs and then have identical things here. Right. So I don't even want to change anything here. It's going to be just identical. So minimize the scope. And then, yeah, and then that's it for closed. And then are we, so we're showing the counts for all of the ethics, but you're saying that we, at least for this issue, we should not show for open and closed. That's because you think it can be more work from the back end side or. Exactly. Exactly. So, so if we time, if we have time, I think we should totally do it and then we should have the issues for it. And we can talk about here, but like just the minimal thing I can think of is, is that, and then we can, you know, we can clean it up in subsequent issues, not, not even. I actually think it might not be more work. Okay. But I need to confirm it. Right. I mean, yeah, I mean, I think it should at least be more work from the front end perspective just to literally code that unless you're, unless the front end is so awesome that. Epic issues and merge requests are identical in terms of the tabs and then you just pour things over, which I don't think is the case. So yeah, as far as like having the tabs is concerned, even from the front end side, there isn't much effort because we will be reusing the list part of the page across all the tabs. So that is fine. Okay. So well, I mean, let's just, those are the requirements. And then you can always go above and beyond. Right. So, but let's, let's target for, for that scope. And then as you, as you work into it, you know, just ping the channel and say like, you know, I've already coded this and it's like minimal risk. And, you know, somebody go and then, and then, you know, you update the description or I'll update description. So I just want to set like the baseline and then. Yeah, I'll put it as a, so Yarka, I'll put it as a, like a stretch. Yep. That's the item. Yeah. Issue. So that you remember that when you're looking at the code, if it's possible to have the counts, we have them. If it's feasible for this iteration, we do it in next. Sounds good. Cool. Thank you. So let me stop there and say, like even like ignore this one for now. So if we want to get all this in, is that possible between Yarka and Kushal? You mean until 22nd? Yeah. So starting next, so starting Monday, right? Should it will be the start? Yeah. I think from backend perspective, it is possible because well, It's mostly just this, right? Yeah. The most, the most difficult will be the first part, like implementing the database changes, changing the models and things like that. Once we have like close and reopen epics with buttons, which includes closing them. Oh, once we have the first issue, which means, which means we also have service for closing them. The second to like close opens epics with buttons and we are quick question are quite straightforward. So yeah. But I wouldn't include the last one. I mean the third directions in due 22nd. I'm not sure that. Okay. Kushal, what do you think? Yeah. The sorting part also from the front end side. It isn't much effort, but obviously as Yarka mentioned, like it might require some effort on the backend side. So we can skip a chase that from the due 22nd. For the rest of the items, we're good to go because the shell that we already have, adding a close badge or close or open badges for each epic as well as the button to perform the close and open actions should be easier. So pretty much doable. And then we want to, at the due 22nd, like I talked with Shana a lot about this. We haven't really, I'm very good solution. I think he thinks that it's like, I'm more concerned like I want to get these merged by due 22nd. And then I told him like, I think I told at least Kushal right and whoever was in it. I think Yarka was going to call us down. We said like, Oh, people don't know and stuff like that. So I don't know last time is because of the subject. So let's, let's not change any process. But given that people don't know about due 22nd. And, you know, we're just going to do our best to, to get the code ready and get it ready for review and pink people to review it. You still think all these four are doable. I do 22nd targeting to be merged. You think that's still possible. So basically, okay. So that's the case. Let's move on to the rest of the stuff that we want to do. So that's basically all these ones here. So I think we should, there's, there's a couple of things here. So this is fine. So, so I think we can, I really want to do this. Oh, sorry. This is, is this the one to, this is the direction, right? Because so this will already be done for 11.3. Okay. That's great. So, so I think this is actually lower priority. So I want to clean up some things. So I definitely want this. But actually there's a lot here. So I actually don't know what's higher priority. Oh, or like I don't want to say, I want to get your thoughts on in terms of, I mean, Pedro and Annabelle as well, like what do you think is high priority? And then for, from like a development perspective, what do you think we can finish up really quickly? What's like a bigger one that we should wait? Any thoughts on these? They're all to me like small things that we need to clean up and fix. I would suggest that we can go after the bugs first in the existing features, which are already in production. Like I can see there is one bug related to system nodes where Epic does not link correctly to the current issues. So those are, I think, long standing bugs that we have been postponing across the releases. So we can go after those first before we tackle on to features. Okay. Yeah. I mean, the other ones are, they're not, the other ones are not really new features either. I mean, like, yeah, none of them are new features if you look at it. I mean, this one is barely a new feature, but you're right. Can I just about the quick question to it from issue references? It is not a new feature because it's not only about adding issue to Epic. It's also about referencing issue from Epic description. Exactly. Things like that. And I think we should do it, actually. But also I would like to say that it will include, I'm not saying a lot of work, but a lot of deeper exploration again, because it's inside the Banzai extractor and so on. And also a question, do we want to have auto-complete as part of this issue? This one? Yeah. Probably no. I mean, let's, the Banzai filter thing that you just mentioned, are you saying that's a lot more actual work that you're not confident about? No, no, no, I'm not saying it's a lot extraordinary. It's more that I have to switch my head to that part of the code. Because it's not like normal Epic issue. But I think we should do it pretty soon. So I wouldn't just postpone it. I'm just saying how it is. Also, I don't think that there's front-end work at all. I think it's on the back-end. Okay. So, I mean, okay, that's fine. So what we said is that we want to do this. And currently we don't have, we don't have auto-correct. So I can't, what is it, %7? So we have nothing here, you said? No, no, no. Right. And I think it shouldn't be part of this issue. Right, right, right. But you made the point, Yarka, that we don't even have that. So that's, that's actual work. That's why it is not related to, to quick exchange. It is basically related to references. Right. But we need, like, if we wanted a quick action where you type that. Yeah, then we need, then we need. Yeah. And then, and then we type this, then we need it. Because you're saying that right now when we do, like, when we do this, we have that automatically. And we don't even have that for ethics. Yeah, we don't. We don't have that for anything, right? Like, we don't have it on an ethic itself either, right? No, we don't. Okay. So we have that issue. So, oh, we do. Oh, oh, we don't have like autocomplete for referencing issues from ethics. Okay. So it would be a good idea to do that first. And then a second follow up would be to do like a third version of quick action. Right. I think that's what. Yeah. Yeah. Yeah, let me take a to do, to do that to, to create two extra issues, Yarka, to do that. Let me. Yeah. So, so yeah, I'll take it to do, to do that. But so you still want to do this. It's still important. I agree. It's important. So, I mean, these are all important. So. Like, I'm fine with this approximate order, sort direction for ethics and law maps. The name. Yeah, I think this is important. I think the sort of direction might be. Well, are we going to do all of this? I think that's the first question. Yeah. Okay. So there's a lot here. So we're like, yeah, Cushon, Yarka, like how do you feel for the rest of these? Like, how can we do all of them? Should we cut off like two or three or four of them? Yeah, I'm not sure how. If you look at most of the issues individually, then they appear like, I cannot say for the backend, but at least for the front-end effort is not much. Given that we'll be focusing entirely on new 22nd issues, at least until 22nd, until they are merged, we won't be able to focus on these. And if we try to cover all the issues within the release, even although all the issues are small efforts, for the most part, we might end up missing some of those. Yeah, I agree because even when it's small issue, for instance, the review process, you still have to go through it, and it can take a couple of days. So if you have ten issues, if you have ten small issues when you are done in two days for them, then you have ten reviews. So I think we should... Well, it's also faster. It's also faster. But that's not always true, actually. I mean, the whole point is, I'm not disagreeing with we should do less, but I'm just saying that we should have smaller issues. To me, these are good issues because they're all small. But yeah, I'm not saying don't remove. Also, if you have small issues, sometimes you find out during implementation that a small issue is not actually a small issue. It's a big issue. There is always this case. Yeah, well, I mean, the assumption... I would say the independent things that we want to do better at. We want to do better at... We want to make things smaller, and then once they're small, we want our system, we want our process to be really good at working on small issues. And, you know, STLC tells us that that should be the case. But an independent process and thing to retron is how to make sure that the things we think are small issues so that's a different problem, I would say. But given, yeah, so everything we said, I think we all agree on to some level of what I want. And so what you said, Cushal, is perfectly correct, which is, yes, please, right now, the purpose of right now is estimating... We did an estimate of what we want to do within by the 22nd and starting on the 23rd provided it's a Monday and you're going to start working. How much are you going to get done by the 7th of whatever? September, August, September... Let's start with September, October, so November 7th. Yeah, November 7th. So what do you think we should cut out given that this is the approximate order? So what should we remove? So one thing that based on the discussions I had with Mark when we were working on the milestone date integration within the Epic, so there is this issue, a show actual dates within tooltips for milestones in Epic sidebar. So this is something that required some back-end effort from the Mark as well, which is why I believe we postponed it until the future release. So again, it depends on how comfortable the archive is working on the back-end side of it. I think we can cut it out at least for now because it isn't a blocker right now to not push for the actual date because the problem was that we were trying to merge dates between the milestones, if there are multiple milestones involved. So it was a back-end heavy thing rather than front-end heavy. So that is something that we can skip for now at least. Okay, I'm going to move it unless somebody complains. Okay. So let's move this. Okay. So this is gone and we need another feature to refresh automatically, obviously. So I will manually do that. We have a workaround. So now we have five. I thought we had seven. So did somebody remove one already? Yeah, didn't we have seven or something? I don't know. I think we did actually. I don't know what happened. Yeah, I think something happened. Does this work? I don't know what happened. Well, we'll find out. We had the sort. The sort disappeared, right? Yeah, exactly. The sort disappeared. That's weird. So what was it called? Sort something, right? Sort epic. This one, right? Yeah, this happened before. I don't know. That's going to be really, really hard to reproduce that bug. Yeah. Okay. So, oops. So, yeah, let's keep it here. Okay. It's still not here. I don't know what happened. Oh, yeah. No, this happened. Look at this. The label. Yeah. How did you do that? Oh, no, no, no, I know. I moved it to do 20 seconds. No, you didn't. No, I did that first. I did do that. I did do that 12 minutes ago. Oh, okay. That issue was here. It was here. And then I moved it manually here after, after we said that, oh, you know, it's going to take forever. To work on. So. Oh, I recall, like, when we were working on the assignee's list, it was a problem. Like if we would move a particular issue from, you know, it was a problem. Like if we would move a particular issue from a label list into open or closed category, then we were supposed to remove all the labels it had. So that was kind of a intended behavior. Not sure if we would want it given that how it turned out to be. That's the intended behavior. I don't think so. I don't think that was supposed to do that. That's definitely a bug. Yeah. Are you, are you just talking about something with assignee lists, Kushal? Yes. So like while working on assignee's list, we were supposed to fix a problem where when to retain the labels and when not to. So for example, if there is an assignee list within this particular board and if you drag an issue from say due 22nd list to a particular assignee's list labels and everything are retained. Just that assignee would be changed for that particular issue. Right. So if you move the issue from labels list into the open category, then what happens is that assignee's are removed and also the label, but I'll have to debug this. Okay. Yeah. No, I mean, I mean, there's, there's, I'll try to make it, I'll try to like use some test project and make the super, I mean, super reproducible. But I've seen this like this is the second time I've seen this. So it's definitely related. And we have clear docs on what we're the, the, the, what's supposed to happen in the case that you referenced Kush also, I'm not worried about that as well. So, but let me write down. I don't have my computer or, so what was the thing I'm supposed to do? So the, so I'm supposed to log this bug and with labels and then also, what do we say? We said something about the quick action. Yeah. Quick action. And, and not labels, quick action and references, right? Yep. And auto complete. Okay. Okay. So that's great. So we, I think everything is more or less. So. Sort, sort was disappeared and is now back. That's great. So what else should we remove here? So these, these like our, this is back end. That's great. So this is back in and front end. So should we, should we try to do ones that are, that don't have too many both back and in front end. Right. So that makes it easier. Right. What do you think? So what else should we kick out this one? Sort direction. From back end perspective, it would make sense. Yeah. I would feel more comfortable. Okay. So let me, let me change this. Okay. And then. So now we have back to five. So everything. They don't look age. Um, Yeah, I think that's fine. Does this look good? Cushall as well. Cushall. Is this look good? You're a mute. Yep. Yep. Okay. So let's, let's go with this. Um, Yeah. Like I want to finish everything. I want to try to like get a hundred percent. Um, I know the dude, 20 seconds are annoying. So. I definitely want all these to be done by the seventh. I don't, I don't think there's any. Problem with that. If we're, if we're like estimating this. I want like a goal to be like, we should have like all of these. In review preferably merge because that's like what we can control. So if you. This is my, your last chance to even kick out one more or more, you know, like as many as you want. Right. So I want you to be super conservative because I'd rather we be conservative and then start hitting a hundred percent and then slowly creep up. Um, I just like to know what, what's the second one, the epic attribution. Um, Um, Um, Um, Um, Um, There's a second one to Epic Attribute showing an issue. Cyber incorrect place because, Okay. This is, um, This thing here in the issue should not be here. If you don't have, Um, If you don't have. Yeah, I actually think it's only front end but Yeah. So this shouldn't be here. Oh, Okay. Because this, this is not this group. a group does not have, um, this group is interesting because, uh, when you have a project, which is not part of a group, you have it in your personal, um, personal name space, then you see the, the message, this feature is locked up great plan, right? That is definitely a front end bug. So the, the, the problem here is that, uh, so what happens is that it, every particular item within the sidebar, uh, is, is consistent across the app. And, uh, the feature flags, especially based on what particular GitLab plan user or group is on, feature flags are only available on Hamilton's side and not the JavaScript side. So this is more of a engineering discussion that even in the front end team, we had, like, whether it is wise to have those feature flags on the JS side, because at the same time, having those on the JS side is a problem because anyone can put a break point within the browser and just toggle the flag and, uh, we'll turn on the feature. For the most part, it won't work, even if you do that, because it would require a functional back end to work with it. So as you can see, like right now Epic shows none because even if user manages to show that sidebar item, uh, it won't be working without the back end code, which has those feature flags check, uh, all the places. So I'll look at it, uh, like how we can fix it. But yeah, it is definitely a front end problem. We need to hide that entire item. Okay. No question is, do we want to hide it or do we want to see the message that it's not available? I think we only hide it for different ways. So again, coming back to that assignee's list feature. So what we do is that if you open, uh, GitLab Community Edition, then in the, uh, add list drop down, we do not show the assignee tab at all. We only show labels list. We do not say like that tab is available, but you need to operate to a different plan. We don't do that. So I think hiding the option would be a better option. Actually, uh, then we need to fix it also for the, first of all, we need to fix it also for the, um, personal projects. And second of all, I think that, uh, we are showing it because you won't use us to upgrade. Right. So, so there's, let's open that. Let's see what, what I wrote here. Uh, right. So are you saying that we should show the, yeah. Can I share my screen so that you can see what I see in my personal project? Sure. Yeah, here. So you can see that in Epix, I have, it's locked upgrade plan. Yes. And so that's, that's what was in my personal, right. I remember seeing that before I wrote that somewhere. Cause that's correct. Right. So that, that's correct. Yeah. Nothing. It's good. And I think if we decide, uh, if we want to fix that, but we should follow this message. Right. And then so you're seeing that. Do you mind, um, screenshotting that, that thing in their sidebar in here? Yeah, that's fine. Okay. So let me write a comment that you're going to, uh, uh, should not show that with no dismissal. Yeah. I already wrote that actually with the right, um, issue sidebar of any issue in that group should not show the epic attribute with no dismissible upgrade UI, uh, only for gold ultimate should the epic. Okay. Yeah. So, so the fix, the fix should, uh, uh, introduce the dismissible, is it dismissible? Um, no, it is not. Okay. Okay. Let me know what I can just quick. Oh, don't, don't show me this again. It is actually. Yeah. It is dismissible to be honest. Introduce that dismissible upgrade UI, uh, for free core bronze. Right. So there's, there's actually, I mean, there's a lot of cases here. So it's a bit annoying. Um, so, so Kushai, this is like, right, like where I wrote here, right, uh, including board. So, so like all the issues sidebars and then the board sidebars should have the dismissible UI. And then you don't mind the description here include that. Yeah, I will. The screenshot. Right. Um, but yeah, that, that, that's it. Okay. Uh, so yeah, so this, this, there is some, some work here, right? Kush, how there's, there are some like, yeah, messy work because if he, if he want the entire interactive UI to make the sidebar option dismissible with the message and all right, but it's already done. Right. I mean, like, like it's already, it's not done. I mean, yes, it is done. But, uh, as, as we had this discussion, like, uh, the sidebar implementations across the app are in So when you look at the sidebar that we show in case of issue boards, that is entirely implemented in view, but if you go to any, uh, issue itself, uh, on, if you visit any issue page itself, then this implement is done, uh, in Hamel and plain JS, uh, in combination. It is not implemented in view. Okay. So we would essentially have to duplicate, uh, our efforts anyway. Okay. Um, let's, let's make it two issues then. Can, can we do that for show? Yep. Okay. Um, so it's already done for the issue page because it's in the, um, it's already done. I don't want to like minimize it. Um, for issue page, uh, so I'm not using sidebars on an issue that will be in. So follow up on this issue to do exactly the same thing, but for board sidebars, because the issue sidebar in boards in view, uh, so, um, that's fine. Okay. So that, that's really, so that needs to be fixed at some point. Um, so what, what do I have to do here? Uh, I think that's good enough, right now. Okay. So let's, let's, let's worry about it. Five and let's fix one thing at a time. So, okay, that's great. Um, what, what, what do we need to look at? Um, so, okay. So we said, okay. So we could do all this. Anything else before we move on to the next topic, which is going to be a quick one before we leave. No, I think we are fine. Good, good with all this. So yeah. So, so focus is on these ones for the next couple of weeks and then, and then work on these ones, I guess, I don't, I don't have any particular order. I think let's try to get them all done, but, um, you know, pick, pick the one you think we can do as soon as you can. We'll go from there. So let's look at our agenda and see if anybody put anything else, you know, um, so, um, if you look at our roadmap, um, this is probably going to get stretched out a little bit. That's fine. Um, we're wrapping up like these things. So if you look at until the end of the year, uh, more or less, there's, we're not doing anything new, new. I mean, like closing is, I guess, pretty new. So, I mean, I'm going to ask like new things once closing is done. But if you click on this, all these things, like for example, this one, I think everything is, um, I don't know. There's a lot of stuff to be done. We released something. I did not know that. This is great. We, today's the fourth. Yeah, but we release, we have new release princesses, I suppose. So yeah, so I'm really confused. This is the, this is crazy. I mean, crazy good, crazy. So we are on 11 three now. That's nuts on the fourth. Oh my goodness. Wow. Um, okay. So I'm going to go crazy and just change it. That's amazing. Yeah, I'm going to, I'm going to change all these really quickly. So this will directly look different. Like we're not used to this, right? Yeah, I'm like, whoa. I'm like, I'm like, what am I looking at? Is this production? So, um, but if you look at me, so. This is these two are going to be done really soon. These are big things like to be done, um, by the end of the year. So that's, that's a make effects easier to use. That's, you know, by the end of the year, um, this we should be done. This is more or less done after, right? So yeah, so you can see, like there's a couple more things to wrap up in the following situations. Um, but, you know, they're not like big, big new things. Um, so hopefully you can appreciate that. Like if you click on the ones here, it's more, it's my estimation, but I don't think it's ridiculous. Um, and feel free to comment if you think it is ridiculous in terms of estimates of like, um, what we can do in the next couple of months or like until the end of the year, because we're in September at any rate. Um, so this one where we're also pretty much done, right? Uh, you need to do these things. But yeah, so this is more or less done. So that's great. So you can see, like in the new year, what we want to do is I really want to do this. So this would probably be pushed back a bit. So I think we should be able to just move left or right. So we might delay that. Don't do that. I think, isn't it? Uh, I, there, there might be some crazy thing back in because like, if you move left, the right or upper down, you might have to load in some new epics. Oh yeah, that's true. Yeah. So this, this might get a little bit hairy. So I don't know, like right now, this is good enough for us, for example, like we're using road maps right now, or at least I'm showing it to you. And then, you know, I don't care. Well, I mean, I do care what's in Q4 2019 and present 20, but right now this seems like it's good enough. So there might be a, a, an argument to push us out, but I think we should do it. But we can, we can argue about that. Um, but if you look at the rest of the, what, what I want to do is, um, basically flexible work breakdown structure. So there's two big areas in portfolio management that our customers are telling us that they're super Uber important. And one of them is, uh, work breakdown structure. And what that means is that just getting more levels of scope, right? Whether that's sub issues or sub epics, or we're not calling it sub epics, we're just calling it epic relationships. Um, because apparently that was confusing. Um, when I was trying to explain, explain it, so I think this is a much better name that, um, uh, I think Annabelle and Sean, at least agreed with, and basically having epics of epics, right? And so that's something that our customers really want, or that's one of the two things that customer wants, which is having epics of epics or having issues of issues and just having a structure downward and then also having capacity management. So saying things like, um, you know, you have, you know, 20 points of capacity for this milestone, you know, somebody goes on vacation, uh, so various things and you can do better predictions and just plan stuff in the future based on that. And that's a lot harder problem to solve in that it's more very specific to teams and, and customers are seeing what they want work breakdown structure first before capacity management. And so if you look at the rest, like a bunch of 2019, what I've done is created all these epics. Um, Annabelle has looked at some of them, I hope, or I've shown it to her, at least, excuse me. And, um, and so you can click through to just look at some of these here. But I wanted to show you one particular issue. I think it's, um, is it? Sorry, let me give me a second. I should be able to find it. It's an issue where I laid out all these things. So it's, um, okay. So I'll place this issue in the, um, the agenda that you guys can look at if you want to, you don't have to. Um, but, uh, oops. So there we go. So this issue, um, documents. What, this is the flow of how users would use epics and start planning at a high level. So it's, it's taking the approach of let's, let's look at how people would use get lab or are using get lab right now or would use get lab. And so a lot of these are us thinking what customers do use. A lot of these are customers saying I want this feature. I want that feature. And then us, you know, putting it all together into one flow and saying like, are we capturing all the steps of this flow? And so this is similar to like how we, we like product marketing was talking about idea to production, like saying 2017. And now we talk more about complete DevOps. So, so it's similar to that. Um, but the, the scale or the scope is, is a lot smaller. It's not like entire get lab is just, you know, within portfolio management or within plan. Say, so you can see it's things like product manager or, or engineering manager or somebody at the company says, I want to do a high level strategic initiative. I want to build, you know, portfolio management features in get lab. So, so, and then I want to do like two or three, three high level initiatives. And then so how can you do that right now inside get lab? You can do that by you can create new ethics, you can update descriptions and you can add labels. So this part of the flow is pretty much already covered. You want to set approximate timelines, far away timelines. You can do that in get lab right now, right? So you can set fixed dates and view them on the roadmap. You want to talk with different people in your organization and say like, are those things possible that you, that you want to do far in the future? And so we can do that and get lab, right? So you have, you know, comment threads, you have notifications, you have to do this. Then once you've talked to these people and things are more realistic, you've talked to engineers, you've talked to operations people, you've talked to legal and so these things that are far in the future are really becoming realistic. Then you can actually break them down into real work, right? So you can break down high level ideas into smaller chunks that are scoping out work. So this is the concept of work breakdown structure. So so we precisely don't have this right? And so these are some of the ethics that are scoped out to to work on these things. And then once you have that, then let's estimate the effort on those things. So we have issue weights as a feature, but we don't have, say issue weights integrated into ethics. So this is an ethic of how we could do that. And then once we have these two, I think are pretty interesting. Once we have scoped out work and estimated work, let's so these like the next four steps are more or less in parallel, or like they're more of a cycle than linear, you would say, how can we schedule the work in different sprints like milestones or iterations? So right now we can do that because we have milestone lists inside a board, but you don't really know the epic. So this concept is swim lanes that that Pedro introduced in one of the issues. And so an idea would be, let's put swim lanes in boards. And so we have issues. So we can move issues across milestones, but see them inside across ethics as well. So ethics being the swim lanes. And then once you're in, once you've done that scheduling, and you've done this high level estimation, it's going to appear 2019 Q1. But then, you know, your engineer manager says that's not possible because of these reasons. And you've done the estimates, then you can go from this top down planning mode to a bottoms up planning mode, and which we this is now already a production, right? So you can go from fixed states to from milestone states, and then that's, that's a great way to do it. And then so then now you can look at your, you can look at your epics here, right? You can look at your roadmap view. You can look at your roadmap view and then you can seamlessly know that some things are fixed states. And then the far away ones are probably fixed states in the future may have a feature that shows that and then the ones that are closer are inherited dates. And then you can do, you can view milestones on the roadmap. So even if you're looking at the roadmap view, then you can, you can, you can start planning your milestones. So like, you're supposed to cringe, Pedro, when you're looking at these, because these are not finalized at all. So don't worry about this. But I mean, the whole point is I love them. So yeah, the whole point of me of sharing this to the folks here on the call is not like, we're going to do this, right? It's me vision to this to you. And then it's like, if you have opinion about this, then please, you know, work on it. And obviously it's more of the responsibility of Pedro and animal to, to, to work on this. But, you know, at Gillab, everybody can contribute. So, so the idea here is that whether you're in the board mode, or you're in the roadmap view, you can still look at how things are progressing, both from an epic perspective and a milestone perspective. So from the board perspective, everything is obvious from a milestone perspective, you still get the epic, you get the epic view as well, you don't lose that. And then same thing for the roadmap, when the roadmap view is really good for showing epics, and it's really good for showing real time, like in terms of like months and dates and or weeks and days, but you don't lose milestones if we can get milestones integrated into a roadmap view. So that's the idea of how you can view your plan and track it over time. And then so you can track the progress, whether it's in the epic or in the roadmap, you can, you can do that by based on some of the things I listed here, like by showing weights and stuff like percentage complete and stuff like that. And then once you're done, you can close the epic. And then this thing here, I just wanted to add it here for completion, which I don't like this idea because it looks really ugly right now. But the concept here is that something that we won't, we unlikely will not use, but maybe we will in the future, right? So I think we should, we should say, we shouldn't say stuff that we won't use ever. But the idea is here. So say that right now, I put this epic here that's supposed to start here and end here. And then so you have some organizations will say like, I put this epic here, I planned it here. And this because I'm a huge organization, it will impact you know, like five different departments will impact operations or impact like, say you're like Uber, right? And then you plan a new feature and then you're going to launch it in like 10 different cities. And then that will impact like operations because they have to hire like managers in those cities, they have to, you have to hire drivers, they have to compete with Lyft, right? So if you're not going to finish your feature, you're going to get screwed over. So if you're the development organization inside Uber and you promise this feature and then the operations team knows about that, they're going to make their plans according to your plans, right? So it's really, really crucial that you once you've set, you know, a quote, unquote an epic that with this scope, it doesn't change. And if it does change, you need to track the hell out of that, right? And so how you would track it right now, what we do is you can only pick one, you can pick fixed states, and you can pick like it's not fixed, right? So that it inherit. So say I plan it from here to here, and then it becomes inherited, but it becomes actually delayed, right? So then we don't know what was originally fixed. And so for GitLab, it's not the worst thing in the world because we missed that quote, unquote the original deadline because we just know that we're just extending and that's just the latest source of truth. And that's okay. But for a company like Uber, like say you're here, and your latest state is like your estimates here and like, oh, no, what are you going to do? So are you go, you need to show that to an operations team that you know you're going to be delayed, or you need to react to that by hiring more. And then, you know, you guys are developers and, you know, mythical man, woman month, and you know that doesn't actually work. But the idea is like, from project manager's perspective, you want to know things are going to be delayed. So how do you solve that within GitLab? I don't know. I have an idea which I don't like, which is this. I think it's pretty cool because we'll just build it on top of our existing feature, but I really hate it because it's so ugly, because you just use both. Right now, this is a radio button and you just pick one, right? The idea here is you put both. And so please cringe, Pedro, if you don't like this, I don't like it either. Yeah, I'm sorry to cringe a little bit. Yeah, but you would show both because if you have both end dates, then you can show both here and here, right? So then that's an immediate way to do it. So I think there's like one customer that's asked for it. So this is not something that I don't think we really need right now. And I don't think a lot of customers are asking for it, but I just want this issue to be here and be able to show it to customers and have them comment on it. But you can see that I haven't even I've been put an epic on this at all. So it's not this feature is not, I think. So I think we are already doing it in some way because right now in epics, when you have the fixed and from milestone, if you have conflicting dates, we already tell you, oh, no, we ended up not doing that. But there was this idea. You also don't need to show on the roadmap, like regardless of like, yeah, we don't show it on the roadmap. But what I was trying to say is that at some point we thought about telling you, yeah, telling the user that your fixed dates are not could not be realistic because what you're planning for your milestones is way into time. You're planning for a very short epic. And so it might not go down as you want it to be. It's a great way to find it. Yes. So so I think it's just a matter of us recovering that design and then see how it would have to develop that in the world. Yeah. And that's also something that Jira portfolio already does. Exactly. Yeah. And they do they already tell you that you're, man, you're going to slip this. Like, yeah, yeah, yeah. Like, you're screwed. Completed everything and it's not going to work. Exactly. Exactly. So to me, that's the like that needs to be done. I maybe that's not even a 2019 goal. And then maybe it's a 2020 goal if we need it. Or maybe it's one of those things that it's like crazy that Jira thinks people need it. But it turns out like nobody actually needs it in the industry. So it's hard to say what we'll find out. But then there's this like whole new category of like what if scenario planning, which is like, again, Jira has like you click something like you change like what if you had like twice the number of engineers? Well, what would happen? And then you you click a button and then this boom magically changes based on that. So to me, that's almost related to that or it should be related to that somehow. But it's not not clear. But anyways, I wanted to show you this thing here, which you can always go to and show you this issue, which I linked in the agenda to see the flow. And then so it's you can think of it as a flow. It's like I to P flow, right? Idea to production. Have we filled out all the gaps? And so right now, the big gaps are the work breakdown structure I mentioned. And then some of the really cool things within work breakdown structure are to me would be would be I have like four minutes. I'll give you one more minute because I know it could be too long. But for example, this is really something that's obvious. But to me, it's just it's just really cool that the way we designed this thing here works perfectly. But the idea here is I think this is just word. So it's really confusing. So I think Annabel had like a really awesome diagram that I can explain the feature I had. And I'll put this thing here. So the idea I have here, which hopefully you folks will agree on, is that say you have a set of issues here and then you have a start date and a fix date and it's fixed, right? And then now you make a dynamic. It's just going to inherit the milestone dates of these folks, right? And then so it'll be start date and date here. And so when you have sub ethics or ethics of ethics, this end date or this end date and start date, if you go from left to right, it'll just inherit the start and end dates of these ethics, right? And then this one will inherit the start and end date of these two ethics automatically. So right now our feature will just inherit milestone start and end dates of issues. But my proposal is that the epic will take whatever its children are and then it will take the left and right of those and then the extremes and then it will inherit and that will be so instead of being from milestones, it'll just be like from children and it'll be very generic. So we don't have to invent like a third radio button. It'll just be from children. And so I really think that's awesome because literally no UI has to be changed, like you change your work from milestone to from children. But everything will be consistent, like the roadmap left and right edges will be consistent. Everything in the UI will be consistent. So I'm really excited for once we do ethics of ethics, we can have this this type of feature. So let me pause there and get your really excited ideas. If not, I'll end this call. So any comments or questions based on that? If not, then we will meet, I guess. I'll comment in the issue, but thank you so much, Victor, for showing this. And it's good that it's recorded so we can like you can trim this and then put it in the channel so that people can look at it as well. Exactly, exactly. So the idea is to we should still have detail like when I joined GitLab and then when I think people new to GitLab are uncomfortable with such details going forward. And like why are we planning all this stuff that we know we'll never do or like we don't have enough resources or people to do as we're not supposed to call them resources. But the idea is that we plan things in a certain level of detail to a granular level of detail so that we can get feedback from people within the organization and outside the organizations. And also we can quickly hire the people or we move people around in teams when we see something strategic. So it is important that there's this level of detail that far in the future, like a whole year. And then we work toward them. I can't find that picture, but that's it. All right, so I'm done talking. If not, or if nothing else, then we will meet next Tuesday. And OK, last thing, I will put this on YouTube. Anything that anybody said that they do not want to put on YouTube and I will not put this on YouTube. So nothing else. Then I will put it on YouTube. OK, talk to you folks next week. See you next week. Bye. Bye.