GitLab Digital Experience, 20 May 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Digital Experience Retro - May 20, 2021

Description

Sprint Retro Doc:
https://docs.google.com/document/d/1kMNiUF2UDuSrMDuzLyRi8OEhVxry_MJoYi38RmmWafY/edit?usp=sharing
Digital Experience handbook page: https://about.gitlab.com/handbook/marketing/inbound-marketing/digital-experience/
Inbound Marketing handbook page: https://about.gitlab.com/handbook/marketing/inbound-marketing/

A

Hello, everyone uh welcome to the digital experience. uh Sprint, retro or iteration retro um today is: may 12th uh we'll be going over the last few weeks of work. uh What what well and what improvements we want to make, um and I think up first- is tyler.

B

Yeah, um I think uh I've seen other people agree with this and talk about it in previous meetings. I really think that we just totally rocked it on these event. Components of the integration and like I just I think we all deserve to like feel very proud about that. uh You know because, like it's, a big one, it wasn't totally planned for we like started this iteration we're like oh right.

B

We gotta like carve out time for this, and then we did and like I feel like we did that and also met some like unexpected uh things as well and like resolve them, and I think we're like really tracking like if we keep tracking at that sort of pace and like problem solving, I think that we will be in a good place, and I just think that we should all feel really proud about that, um and I appreciate all the support that, like everyone has uh been giving towards that effort, because it's a big one and it's also yeah it's great.

B

So thanks. Everyone and good job uh laura.

A

uh Yeah um we had a meeting this morning uh myself, just stephen brandon michael has, I guess, decided to kind of split up the 15 pages, okay, that um stephen and brandon were leading uh and kind of divvying up some of that work. So just- and I can take it on um because our okr has shrunk a little bit in scope. um So having that kind of split uh feels good feels like we're being a little bit hurt. I think we were a little concerned that you know. 15 pages was a lot.

A

We weren't sure how much how far we'd be able to get on all of them. um So this should help divide up the work but yeah good collaboration, and then uh stephen is next.

C

I'm gonna say the exact same thing more or less in a nutshell, and was good to just collaborate. Break things up feels a bit better distributed just a bit more manageable and more attainable, and I think at the start it just looked like one big giant hill that we had to carry to climb over three months and it didn't really feel like there was actually any path or plan. Now there does and so yeah feeling a bit more comfortable with that. So very happy.

A

Sweet um and then we have a few things to improve on, I think lauren is up. First.

D

uh Yeah some high priority items crept in this iteration and like the unfiltered blog and then renaming marketing to uncategorized are just examples. um I think this is gonna increase throughout the year. So it's not like this is gonna stop happening. I think it's gonna happen more.

D

So what do we want to do as a team, so we reserve capacity for this, or do we just de-prioritize when it happens.

B

um I felt and I've seen laura's typing: do you want to go first, laura, as you type.

A

uh Yeah, I was just saying that I think um like lauren and brandon are known to be on this team. So I think if there are problems like a lot of times, they take on the brunt of it, because it's just like someone reaches out to them directly like I had no idea. This was even happening, I'm in my own little world like flowers and sunshine, building my components so yeah. I don't know how to like holler that like we can help too. You know.

B

um I think that segways nicely into what my thoughts were for this is like yeah over the past, like three or four iterations, like I realized that I was like over committing myself each time, and so this last one I under committed myself and I'm like. Okay, I'm like we talked about like event components, and I was like all right. That's like the top level thing I have to do so.

B

I like took on three or four of those and then like one additional thing that like was just a just, do it it's a quick thing, and but I knew that I had more capacity, and I was like okay, like I'll, probably just pick up more of the event components and then like the unfiltered change came through and instead of picking up an event component and I did drop the like other one or I'm like pushing it back to tomorrow instead.

B

But I was like oh cool, like I planned for this by under committing right and, like I don't know, if that's I don't know, if that's systematically going to work but like we have a note in our planning document to like you know, be um conscientious of like actual capacity and, like you know, and and plan for the unplanned, and I think, like I found that this iteration, like you know honestly like doing the unfiltered block, did not change any of my deliverables for the commit event like zero impact for me, and I feel really good about that and that's like specifically because, like I committed to like, I gave myself that time that I anticipated, I would do a little extra on commit, but it turned out that I serviced like some other urgent thing instead, so that worked out super.

B

Well, it's not very systematic because, like part of that was just me getting lucky, um so I don't know if, like you know, I I don't know, if that's actually the I don't know what the like systematic like process answer is, but I do know that, like our directive that we've been trying to, I think we say every time in planning of like hey, give yourself more time than you think you need.

B

I think that like gives us some reserve capacity, and I know that brandon is also doing a lot of good, like actual sprint point math to say, like here's, how much work we can actually do and like I know that he's accounting for some of these unaccountable situations but yeah, I don't know um just being mindful. I I don't have a better answer. I see uh just has stuff written down.

A

Yeah and just to add to that, because that's a great point but um and lauren already said this like de-prioritize something, but I think also it's important to for whoever's asking the thing to to make sure they understand that you have to de-prioritize something and be like okay, we can do this new change a, but that means we can't do change b, like you have to kind of pick and choose, um and they might be okay with that they're like yeah, a is really important. Let's do that or they might like, be like oh shoot.

A

We really needed to do b. Also so, like, let's stick with our original plan, I've seen it go both ways. It just depends on what the thing is, um and just I didn't write it here, but I was just thinking too, like maybe next quarter, we're still iterating on how we do our okrs, but something we can also think about is being better about both waiting and like stack ordering the importance of things so that we know.

A

If something comes up um and it's you know a weight of three, we can like look at what else is a weight of three in our backlog and is like the least important and maybe like take that out. So something for the future.

A

I know that, um like sometimes things come like if something small comes up on like the is it known, slack channel or something um if it's related to our team, a lot of times like one of us will see it or michael will tag like at digital experience and we'll kind of pop on it. Is there any way to like leverage like if someone gets pulled into something that they're really not sure they can take on to like hand it off to one of us?

A

I don't know, I feel, like the unfiltered stuff had to be like a full stack person like it's. I don't know.

D

I don't know I would. I would recommend, if you're unsure about something to just um go through michael and ask him to prioritize for you. um I like the idea of under committing and then having that project that you know if you do have more capacity to know. Okay, look I I can bust through this and add some more like components.

D

We likely have that event project right now, where it's easy to like devote extra time to, um and maybe we'll always have a project like that that we can always work on.

D

Or tech that you can always work on tac, that.

E

I don't know I, I wrote a point really quickly that I, like I'm terrible at time estimation, but I just feel like there's like a root problem that still like has to be figured out. You know like a root problem and a root solution, because it's like, I don't think it's reasonable for you all to be like.

E

Let me reserve time in anticipation for like things that come up in the middle of the iteration at all times. I feel, like that's a little unfair to you all personally,.

D

I agree with that, but when these things are coming up, it's like we need it done tomorrow. We need it done in like next two days. um I don't know what the solution is there, but I think it's going to keep happening.

B

Yeah, I think that, like my outlook on that has been like, I I I I think you have like a really. I think it's a really good point and that's like- and I like, I wish that, like I have ever worked in a place like that and like I don't mean to like, be dismissive, because I think that that is like the right like the right way to work is to be like really clear about scope and like boundaries and like have a good sense of like what you're gonna do.

B

But I also feel like it's um like there's a place. We live in right now, where we can try and we can try and get better at that and honestly, like your lab, isn't too terrible at that like compared to other places. I've seen and I think, there's like there's a lot of value for us as like collaborators to like be flexible, and I I think that you know.

B

Strategies like under committing at the outset of each iteration are ways that, like we can be highly collaborative with other teams who who don't have like a good venue because, like some of these things are like it would be reasonable if they came into our planning process at the beginning, but either they don't know which is reasonable because they have their own processes to know or like they couldn't know, because they couldn't know that they were going to need, like you know, at the end, like ever like we're a few levels away from the actual unexpected pressure right like because I think this.

B

I think this would be different if we had. You know like if michael was requesting mid, mid iteration every single time like hey, here's, some unplanned work but like we are, you know, collaborating with folks who are working with different pressures, and I think it. I think it makes us really valuable if we're able to be flexible. But but I know just it's, it's like a balancing act and I again it's not that's, not a systematic answer and I also feel like.

B

I don't want to dismiss your really good point of like it's not great, to have to like always be ready for unexpected work, but like it feels good if you get it right, because people like it and if you set your own expectations about like what you think you're gonna get done, because you know that like either something will come up or nothing does and you can do extra like, like. Maybe that feels good for your sense of productivity too. So just a just a counter point to what I think is your very reasonable perspective.

E

Okay, I think you have the next point. Tyler.

B

Cool um super small one uh not like huge, like I just I noticed uh our ruby version or the dot ruby version file which just like informs some like automatic, like ruby version managers of like what version to use and like our ci pipeline um unrelated mr came through, and it like rolled it back. I left a comment for the person who made it.

B

My guess is that, like they had some issues with the 3.0 migration and there- and there was a quick fix for them, that 2.6.6 worked uh and so they're, like oh cool I'll, just do this and it works, and that's a super reasonable thing to do. If you like, are you know not working with that every single day um and it's a quick fix to roll it back to 3.0, but um it's also like some unnecessary.

B

It's not very efficient, like we shouldn't have to do that like we should set the ruby version and it shouldn't change until people who like need to set the version, wanna update it, um and so I mentioned note here about like can we use like I, I've never used the code owners feature, and I remember a cop someone said somewhere that, like it doesn't work reliably or maybe I misheard something so I'm wondering like.

B

Can we use code owners on p like and like this would be good for, like our webpack config and like our middleman, config and like ruby, were like things that impact our day-to-day, that someone might reasonably say, like oh cool I'll, just like edit this to make my thing work um and not like see the cascading challenges that happen afterwards, anyways um barker, you mentioned, there's a related comment: um are we able to use code owners? Can we do that? For this? I I.

D

Are we using it in slippers like you, can't merge something until it gets an approval, or is that a repository ride.

B

I think that's repository, I don't think it's like per file, so I think so. I think the challenge like with this the person who rolled our ruby version back um like is a.

C

B

So they can just change anything which is super fine, generally speaking, but like it also means that nothing came up. That was like hey, like here's, a group of people that are impacted by the ruby version file right and I think that's that's. The thing is like we do have a lot of maintainers who are not working day to day with the things that, like sometimes they accidentally or unknowingly, change.

D

um So that that came up during the initial discussion of the unfiltered blog and that's the issue, I commented there about permissions and workflows and um if it would be possible to set like a group of people um as approvers for mrs in certain areas of the repo, uh I think it would be beneficial for us to look into this and research, the the code owners and all of the settings and features that gitlab has. um Because there's that's gonna, increasingly be something that we're gonna need to know about. Yeah.

B

For sure, okay, um I've made an action item for myself uh to do that at the beginning of next iteration, I'm taking a vacation second week, so I've got a short week and I can do some of these like miscellany tasks. So thanks.

A

Cool, I think that's it for retro. Thank you so much for joining.