GitLab Digital Experience, 25 Feb 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Digital Experience Retro - Feb 25, 2021

Description

Iteration Retro Doc: https://docs.google.com/document/d/1kMNiUF2UDuSrMDuzLyRi8OEhVxry_
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, um it's thursday february 25th 2021.. This is the uh sprint retro for the uh digital experience team's latest iteration, uh which ran through the 16th of february to today again the 25th. um So we can kick things off with uh things that went well. um I've got the first item here, uh which is a quick one. We had our ui to code working session on uh this past monday and I think that, like I walked away feeling like it had been massively productive like a really good way to like align.

A

I think you know if we're thinking about sort of the, um what is it uh forming storming norming performing uh model of like team dynamics, um michael talks about it uh every now, and then I think it's useful. um It feels like that meeting was a really great part of like norming right, like I felt like we got very aligned on it, so um yeah, I felt good. I see a couple. People have notes there so feel free to hop in on it um or if not, we can uh jump over to jess.

B

Let's move about to just because I'll be saying almost the same thing you just said: yeah the ui to code was really really productive. We were able to discuss about things happening between the designers and developers, which was really amazing. I actually told uh michael during my one-on-one, that we should always do this almost every single week, because I enjoyed it. So that's a very big plus. I love it over to jess.

C

um I I probably should have put eight before seven. I want to shout out steven um for kind of taking over the featured partner template page design, which totally freed me up to focus solely on research for pretty much the last week and a half, and it's just been like very enlightening to see how things tested out and then also just kind of I'm, not a researcher.

C

But I I've done some and it's been really like eye-opening and it's like how much work it really is to do research and how much it would benefit benefit us to have a full-time researcher on our team. um Who could help us out with this stuff. So um it was just yeah very eye-opening week and a half for me and fact is on me.

B

Oh, thank you. So uh this last string there was a lot of collaboration between me and jesse lawrence. You know everybody that was totally amazing.

B

That really went out well, especially with jesse, who are going back and forth on the okay moving things around, and I love how we all pay attention to details, and I really really like that that collaboration is awesome and thanks to a big shout out to brandon after I submitted my release, we found out that there was a z index that was affecting one of our pop-ups, so it quickly took the best decision to take a good action on that. So those are some of the things that I really saw that went well last week.

B

I think we are doing a very good job on collaboration. That is awesome and I want us to see. I really want us to keep doing that so that we can always get the best results on every project that we work on. So that is that for collaboration, and the next thing is the migration of the uh child topics. It was really smooth uh so shout out to tyler and lauren for putting that stuff together. That's really really! I really like that.

B

So I will still be submitting some issues about what we need to improve on on the uh on the neglify cms about things. We can just track around to make things much more easier and better so over to tyler yeah.

A

Yeah something else that worked really well. I felt super good um thanks again uh to brandon um and becky to uh for, like clearing up a lot of things for me about gillab issue boards and also, and specifically brandon, for like putting together issue parts for us. um That was like super helpful and I feel good because um one I enjoyed using them. um It was nice to like be in the tool right and then to like from a bigger picture.

A

It felt good to like be actually contributing to this, like dog fooding ideology that I know gitlab is trying to um promote, and I know that there are like challenges there so um and like it's, you know it's not perfect and I've got things to improve on for it, but like overall, I was really pleased this iteration.

A

Like I didn't open my. I didn't open the sauna once in this last in this last week, I think earlier in the iteration I was but like yeah, I figured out how to just manage my work through git lab and uh up next is brandon.

D

Yeah just along those lines, uh I'm I'm really happy that the new sprint and kanban boards- uh sorry, iteration and kanban boards- are helping us out uh this week. So um yeah I'm glad I could get that to y'all and I found myself using it and I'm glad we could align with becky on kind of the use case for it and for our internal team. uh It's been helpful, so yeah. I think that went well over to tina.

E

uh Thanks brendan, I just want to say that I wasn't using boards before you created the new format for it. I would open up my boards and immediately close them out of panic. So thank you for that. I enjoy using them now.

E

Another thing that went well is the collaboration with the team on the block templates, which actually happened during the ui to code working session.

E

Unexpectedly helped me make like some big decisions, including uh including it into slippers, which I hadn't even thought about, and a special thanks to jess, who is so quick um whenever I need feedback or uh not just on this project, but on everything that I work on. She's really quick. To give me a feedback and she's uh usually always right and sees things that I don't see on to lauren.

C

uh Yeah I just wanted to uh shut up. There were a few times this week that I was like ran in. Did I break the merge train and he was super quick to like you know, look into it and fix it like brandon. Do you see a million errors on the homepage? You know it was a a quick. You know turnaround. Always you know if it feels very supportive and I feel like anyone on this team. I could like be like help and you guys would come running.

C

So I think we got a good team dynamic, but specifically I leaned on brandon a lot this week. So thank you. I guess we can move on to things to improve on. I think tyler is the first point. There.

A

Yeah, um I totally like so I I said scoping and time estimation, which is probably part of this, but I think maybe there's another piece that is really so I I fell behind. I bumped like one I was overly ambitious. um I didn't appropriately take into account the shortened iteration we don't have tomorrow to work on. I think that if I had all day tomorrow for focus time, I might have bumped fewer of my issues back.

A

I think I still would have had to, but anyway I bumped issues back to the backlog, and I think that I'm not going to like deliver the blog template like 100, in the way that I'd like like wish, I could have done fully to like rise up to that, um and so part of that is like a big thing and like it is too late now to be saying that basically, for the first time today like I feel like that, like that's a failure on my part, but I also something for us to improve on collectively um is like like who do I get like who should I have told right like one?

A

I should have realized this. Like by monday or tuesday, but two it's like I even if I had I don't know who I necessarily would have gone to, and so you know, I think, there's a part of like you know, agile or like scrum of like daily standups. I don't know if we need to do a daily anything or like add more to the process.

A

I know we're trying to like not change the process too much, but um something we can't improve on is just like, maybe some more frequent check-ins or a clearer process of like here's. Here's what to do when things seem to be off track, because I think I just ended up off track and didn't know what to do about it.

E

Tyler, I'm going to jump in there. I think that um michael had created a bot, um a check-in bot. I don't know I I stopped using it. I I guess I didn't.

A

Win it anyway, it went away, I think it stopped working, because it was like tied. It was tied.

E

A

One, the channels that got deleted. That's.

E

What happened so because that's that asks specifically how you feel who do you need um support from so maybe we should talk about getting. We.

A

E

In action, I'm.

A

Gonna drop an action item just to like uh fix the iteration bot, because I think that I think that that would have at the very like that would have been a really good opportunity for me to like. Do the to like raise my hand about this earlier on.

A

Okay, um so I'd be an action item there and I will figure that out and we can tie it to whatever the appropriate channel is. um But another.

B

A

To improve on- and this is just me like um and brandon's- got an issue on uh or an action item about this to you about, like weighing issues I think, like I didn't, do a good job like I said, yeah.

B

A

Hello I'll do this thing and I didn't look at it before. I said that, so I.

B

A

Didn't have a good idea of the scope of things and like had I done that much earlier, um I probably would have like had some feedback about producing the scope or even just said, like I don't have capacity to do the whole thing et cetera, et cetera, so um anyways, that's on me, um but I think there's a couple: quick solutions for slackbot fixes and um brandon suggestions about like doing weighing and stuff. So uh I I apologize to anyone who's um workflow.

A

Either people who are watching are people who are here that I've impacted on for this one. So sorry about that.

A

I think you're next, you know.

E

Oh I'm next, um we've already discussed this, but um file handoff communication could have gone a little bit. Better um lots of information was um lost in fragmented issues, an issue being for a content, then an issue being for wireframe and then an issue being for design and then an issue being for engineering and really important information got lost on its way.

E

So um there is an issue, an action item, to work on that process, but um there's definitely room for improvement there and maybe there's something that we can do as a team in the interim to just better communicate key things as they come up.

E

F

Sweet so becky. I hope this is the right thing. I believe this is part of it, but it was not the thing I was looking for. It's okay, um becky asked for feedback on like the the process in the way that we do things. um I believe this.

F

She asked this uh tues monday or tuesday this week um and it's an issue and when someone asked for feedback for me for things that, like weren't working, obviously like, naturally, especially being myself, I wrote like an essay, and I made buckets into like agency day things that aren't working like the way that we structure our two-week time boxes like uh and all of that related to kind of like scoping in time estimations.

F

I think that they're, the big thing that I took out of that that like is as I've like, sat through it like you know, through a couple days and thought about things. I made some recommendations there now that I, even after thinking about it, just doesn't make sense but like the one that still feels like would be good for us is a refinement process and what I mean by that is helping break down.

F

Like issues before we get them, and that will help with like the weighing process and trying to figure out hey like we have a cue before a sprint before an iteration starts like we have a list of issues that we can pull from, that we know have been scoped appropriately for our work. That way we're not taking an issue. That's just like wait like 20 for something that's like taking us, not that much time.

F

So that's something that I think we have to improve on and so yeah and by the way, I'm just having a pop-up that my camera battery is low. That's why I froze.

F

But yeah that's something that I just wanted to bring up just because I thought that that is helpful and I can link the actual issue in which I wrote all of these comments down up next is tyler.

A

Yeah, this is a follow-up on something that went well, which was uh dog fooding issue boards, um things to call out that I learned even in just like I was like okay this week, I'm only gonna manage my work through the gitlab interface and for the most part that works pretty well.

A

There are two to me like the two biggest things that come up are one like for me, like I'm, a big like to do list person and the to do's in the gitlab interface are like just your mentions and stuff and like that's not what I need. I like, I need just a free form like I just need a to-do app like anyone.

A

You know like a thing that, like you might do if you were learning a new programming language like build a to-do list in a thing, I just need that and I just need it inside of get lab. I filed a product suggestion uh for it. um You know- and I think that's the virtue of dog fooding is. I was like hey. You know this could improve the product, you know and I'll just kick it over the fence and and see what priority it is um for now.

A

um What I'm doing is um and I'll link this. um I actually have like a little like work notes uh like a work notes repo, um so you can like. I think I stole this idea from brandon. I don't know what his looks like, um but like it's literally just markdown files, um and I just like- and so but now I can do it all through gitlab, because it's like text editor markdown file, like I just keep a like stream of thought and you can see my commit messages.

A

Are tara just wrap the day start the day like whatever, um but I am now like. You know it keeps me fully in gitlab right like I could even manage this through gitlab, if I really wanted to, but I prefer vs code for it. um So that's been useful, but it's still not per like. I can't drag it right. I want to like drag it. I want to check off the things like I want to do. I wanted to do list that I want that.

A

So bad, um so that's missing for me in issue boards um and then, mrs. I think that so the to do's are like a personal preference thing and like I could live without them, because you, low tech, just have a notebook whatever. um But I think that mrs actually, like the really the biggest thing, that this is like things we can improve on. But maybe we can't I don't know like. Maybe we need to escalate this or something it.

A

It feels silly to me in so many ways and I've read through some of the issues about like what the actual technical challenges are, but like it feels silly that I can't get my mrs to show up in the same place as the issues like I just because that's where so much of my work happens and it's in and the big thing and like it's like okay, a lot of mrs are related to issues, especially if it's something where I was assigned an issue, and then I made the mr I can like link them up appropriately, but very often I will receive like, and so in this last iteration I did like three or four different reviews of community contributions, so I there like wasn't an issue one.

A

It's just like someone's like hey here's the thing and then two like. I have no control over what they do to link it to an issue and so like then I I just lose it like. There was a couple days where I was like. Oh right, I had an mr and it just it, because I was trying so hard to stay in gitlab.

A

I didn't see the thing because it didn't show up in any of the views that I was really looking at, and I know it's there right, if you click in the right hand corner but like if you're trying to um if you're, trying to filter your things down like it's there's, just not an easy way to put it where I personally want it to be, um and the final note on this is that, like there's a whole handbook page about how to work at gitlab- and it's like everything starts with an mr and it seems to me that that is not how it happens.

A

It seems to me that everything starts in an issue and, like the handbook page is very explicit that, like mr should come first and like issues should be like. If an mr is not is like, if you really can't get an mr together like, then you can move into an issue, and maybe the handbook page needs to be updated. Maybe that's just like that. Philosophy has changed, but it feels like the thing that is supposed to be.

A

The first option is like unworkable, and that seems there's a disconnect for me so um but hey this is dog fooding. I feel like really like yeah. I feel good about having come to this. I think other people have expressed the same via slack so uh javi. I think you're up next.

F

I just added this note, because I forgot to talk about this in the feedback issue um in the monster issue that I added one of the big things that I want to work on is not project managing so consciously during the sprint itself, because that means that I deliver less in the sprint, because I'm spending my time trying to figure out if I have the right labels, if they're showing up in the boards properly.

F

If I'm doing this, and so the suggestion that I have that could be an action item is like for us trying to be because before our sprint, we have some time. I think some working time where all of us are like awake and should be working. I think, before our sprint planning meeting. What I suggest is, after that hour, is over.

F

Let's try to commit to changing our plans as little as possible that way, so that, when the sprint planning is over we're locked in for the sprint until the end of the planning, until the the retro session that we have right now. So that's kind of like what I have in mind. I don't know how other people feel about this, but I think that it's kind of silly for us to be managing our work. While the work is happening because it slows us down.

A

I think there's probably a balance of like flex that, like I think we should be able to change things when it makes sense to you because, like I, I want to at least be able to accurately label things when things change, so that the like the information is up to date. But I agree with you that, like the I agree with the goal of like change as little as possible about the plan during it- um and I think, like brandon's upcoming action item about like weighing and capacity, will help a lot too.

A

If, like we get better- and you know this is like we have this- like it's labeled retro, eight right. So it's like we're on our eighth iteration of this. uh I think I've only been here for three so, like I think you know time. Estimation is one of the hardest things about like agile, scrum or anything in the world. So I think we'll only get better at that, but, like yeah, I agree with the goal um and I would just say, like let's add some flexibility to re-categorize, uh just to keep information correct.

A

Cool uh we can move to action items unless anyone's got something before that.

A

uh I will kick us off here. um uh Danielle in slack uh had a suggest. We, I like posted, a video about some slipper stuff and she asked if we were going to like release it, um and so I kicked off an issue which has a um quotes from the slack thing. Since slack is not a you know, document of record and uh some links and stuff too um nothing to really talk about.

A

I think I think the action item here for me is, I think, that next iteration planning we should have this teed up as like one of the possible work items and then work asynchronously in that issue to talk about what that looks like and like what what the requirements are and what the value is, and so I just want to make sure that this finds its way into our planning session next week, um even if it's still not the right fit to do yet. um That's all I have for it so um on to brandon.

D

Yeah, so I I apologize, this thought hadn't occurred to me earlier. um You know we're all still kind of forming as a team and forming how we do our uh processes and that sort of thing, but it occurred to me that something we might benefit from uh is weighing our issues uh either.

D

It would be ideal to do it async before the sprint planning, but also while we're doing the sprint planning. We just want to make sure we're not taking on too much work. So it'd be good to kind of go over the weights of issues. uh You know kind of agree. If we, you know uh in agile processes, usually the team gets together and says. I estimate this task is going to take x, number of points and other people on the team might say. Oh you know, did you realize?

D

Also you need to do this, and so it may be more than you think, or some other people would say. Well, we could do it that way, but maybe we can do it this way to reduce some of the cost, and so you know it can be beneficial to weigh things together as a team.

D

So we might want to try that uh in the next uh iteration just kind of planning together how much we think an issue will take to do and that way we can also keep a careful eye on how many points we think we have capacity to do during a sprint, and then we can also retrospect, okay, that took longer than we thought, or we actually have fewer or more capacity than we thought during the sprint. uh Sorry iteration, I keep switching between iteration and sprint, as I'm talking the same sentence uh but anyway.

D

Yes, um so that's kind of a good process. I I hope we can uh start doing on to tyler.

A

Yeah, you answered my first question, which is like how do we handle things we're unclear about, um and I think the answer to that question is first like we do async work before planning session of like putting time estimates or what you know if yeah doing some estimation right- and this is part of like managing the back the backlog overall um and then, like you, said, like the negotiation in the planning of like hey, like I'm gonna work on this, um and I think it's gonna take this like this many points and then sort of the conversations there um yeah.

A

I think that, but that answers my first question is like we should get to clarity by the end of the planning session, either through asynchronous work for clear tasks or some synchronous conversation for like unclear tasks. um But my second question is about prioritization.

A

um I know that, like we have okrs, um we have, you know like we get messages from managers and meetings and stuff about priorities and, like often like you know, I think one of the best things about agency day stuff. Is that, like those come with due dates and like very good reasoning for them, so I think but like for other stuff, I guess you know like like who sets the prioritization right like like?

A

Is there someone else we can go to for what are competing priorities um so far, I've kind of been managing myself and I feel like this is a big thing, because it's it's not just about like the weight and the time estimation it's also like if you've got 10 points and you have 12 points worth of stories or worth of issues or whatever it's like what to get dropped, and I don't know like if, like I don't know how to answer that question or who to go to for it um or if there even is an answer for that at this point.

A

So that's that's my only question about number. That's the most pressing question I think for um as we iterate on planning sessions is like who calls the final prioritization here.

B

Yeah, I think it's all go ahead. Oh sorry, brandon, so I think I ran into that just this week because there's a new uh partners page that I'm supposed to start developing and I'm planning to push it to our next friend. But when I discussed, I think becky sent me a direct message on slack and we got into that conversation that I have something that I'm doing this week.

B

I know this is audience, but how do I prioritize what is important and what is what I can push to and explain and she actually said: okay, let's put this public on slack, then I can comment on this. Then we can get michael involved and we can know how we can move things around. So I think becky is a it's good to go.

B

It's someone we can talk to and if we always put that public so that everybody who has maybe a stake in that project can also contribute that oh, this is not a big deal. This is not what we can prioritize on, because I was actually trying to prioritize on the um google tag manager that is not firing on our youtube. Videos that was what I wanted to prioritize until I put it. Probably then, michael becky, everybody said no, let's push the partner page as a more priority.

B

So I think those are the strategies that I just used this week, but I would love to see what brandon will suggest.

D

Yeah, what I would suggest is basically along those lines. uh My experience so far has been. uh You know we we set out to do x, number of things per sprint, and these are the issues in it. Sorry iteration and uh you know, then uh something comes up and I either have to drop some issues or something's taking longer or something new comes in whatever the case I need to re-prioritize I'll.

D

Take my first guess my gut is that I should be working on x instead of y, um but also I will go into slack and check with becky or whoever's related, whoever the stakeholder is. That says, I need this done or or you know, maybe you know I say oh, I might have this slip.

D

Is this going to be a problem or whatever the the reason is um you know I'll go into slack and ask um becky's usually got a good idea of what's going on um so you know, I tend to start with her uh if you know something deadline related because she's involved in a lot of the deadline, related items, uh you know in terms of agency, day work and whatever else, and you know cross-discipline teams and that sort of thing uh you know and there may be a quarterly okr or something that's competing with it.

D

So you know I'll pull in michael or you know, whoever is available and just say hey. You know I. I think this is the order. I should be working on things. Can you confirm if something else and I'm needing to boot, something out of the iteration cycle? uh Can you confirm that this would be? You know what I should do or do you have a different suggestion or whatnot um so yeah? It's it's basically a sami described it.

D

You know, um jumping slack, ask um and you know yeah uh becky's a good point of contact on that. If, if someone is not available, uh you know because I know we- we had, for example, uh snow in texas recently where becky was unavailable and you know michael was on. uh You know pto, for I don't remember if it was, uh he was moving at that point or if he was doing a ceo shadow or uh you know and danielle's out or whatever, then we can take our best guess.

D

You know converse as a team ask the stakeholders, you know what they think and that sort of thing, um but yeah, there's there's usually someone to ask in getting that group conversation in in transparent public uh is, is a good way to do it and don't feel bad. If, if uh deadlines change or if you know you can't make something that you thought you were going to make the the earlier, you communicate that the better, because then we can all adapt so yeah, definitely good to let people know and ask questions.

C

I think there might also be an opportunity now that we're all going to be operating um as inbound marketing with the same sprint dock. When we go to update what we plan on doing the sprint, um a good practice is just putting it in order from top to bottom of what you you plan on working on in order and then becky's going to be in there looking anyway, and if she sees like something at the bottom that should be at the top of your sprint.

C

Then maybe there's opportunity to like kind of readjust and and communicate that way.

A

Sweet. Thank you.

A

um I'm up next with a quick one unless there's more on this point. um Just actually. I just put this in my to-do's: I'm gonna fix the iteration slackbot update thing like we talked about before, uh so I will get to that uh early next week um because I think it'll be a quick like reconfigure, a slack bot thing and then javi. I think you got the last action item here.

F

Yeah, um I wanted to specifically call this out because I think this will help us with some of the like the weighing in the scoping, like some of the things that brandon has brought up in this previous action item.

F

I'm trying to think about like how this becomes possible without making our planning meeting longer in the original post I had for becky. The simple solution was make our meeting longer because we have more time to do more things. I don't know if, after thinking about it for a couple of days, if that necessarily works for us, I have to think or not I, but we as a team have to think about what is our way of making sure that we kind of have an idea of what's on the horizon. I'm sure things will change.

F

You know as things priorities often change. I don't think we should be planning more than maybe a sprint ahead at any given point or an iteration. I'm sorry it's funny because I suggested the name and I'm the one who's breaking the rule, but trying to figure out how to get that to work within our two-week little time box. So that's something we can make an issue about, I believe, or we can make it like a similar, because it's I think, it's very related to what brandon has listed.

F

So it can all just be like one thing but yeah. I just want to specifically call that out.

F

Anyone else have any last minute things our agenda is done, but yeah and we're three minutes over time, but thank you all for coming. uh It feels natural that I end just because I was the last person speaking, but uh thank you so much. I will see you all soon.