GitLab Digital Experience, 7 Apr 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Digital Experience Retro Video - Apr 7, 2022

Description

Digital Experience Handbook Page: https://about.gitlab.com/handbook/marketing/inbound-marketing/digital-experience/

A

Hi everyone uh welcome to digital experience. Team uh iteration retro today is thursday april 7th. uh We'll go over some of the things that went well and things to improve on from the last two week period. uh First up for things that went well, tyler has the first point.

B

Yeah, um for me, it went poorly, but for parker like save my bacon, um so thank you. This is just praise parker, so we were doing things where we're like. Oh, let's clean up all the stale images from the middleman repository and I felt super clever and I was like I have this clever solution. It'll work great and it did. It was like super fast and, like honestly, it's a great it's.

B

A great thing like I'll tell you about some other time, um but I started from a faulty assumption uh and I ended up like deleting probably like 100 images that people actually needed in the repository. um The thing that barker did, that uh was awesome, uh is uh gitlab sub value. It's like two-way door decisions. I think it's under the iteration value and like I was like yeah, we could like delete like 200 images a day I was like we'll do it in badges, 200 and parker's, like maybe 50, and I was like okay 50.

B

and then, uh which is great, because what I needed to revert those changes um it like took like less than an hour, so we fixed it super fast um and anyways. Thank you. Parker. You did a really good job of containing my cleverness because it was not particularly clever um justin, uh the best this like. I wanted you to come work with me again, because I wanted these memes. uh You all see a funny meme that justin made for me uh in the middle of my frantic reverting commits.

B

So thank you to the both of you.

C

This this was like an instance where the template was super cleverly engineered in itself, so like we just didn't even think about it, and then it happened. We're like oh okay. I see.

B

I do have another clever solution which I'll get to in like may. Maybe so you know we'll we'll revisit, maybe I'll be clever again mateo. I think you're up next.

D

Yes, I just wanted to say that I really like the approach that nathan took for the migration of the pricing page, the creation of the issues where every issue is just one component. Instead of having one issue for the whole page, I felt that it really helped organize the work and tyler. I think you also have one point here: yeah.

B

I just wanna say out loud. uh I agree I think, like nathan has really been holding it down in terms of like splitting stuff up and, like I don't know, it's just like a thousand percent easier to figure out what I need to do every week uh with him doing that. So thanks.

B

So a plus one on the thank you and also wanted to highlight that megan has also been doing similarly as well as I think you brought it up in the red, yellow green section megan that, like we've, been like collaborating really well and like honestly, it's thanks to you like splitting it up in a way, that's easy for me to like jump in and jump out and, like you know, not get in your way right. So, thanks to the both of you for being super smart.

E

I have the next one, and I just wanted to thank uh mateo, john and lauren for helping me out troubleshoot the the issues uh I found that uh you're very helpful and just wanted to verbalize that so thanks.

E

And we go to things to improve on so you're next again,.

B

uh Yeah this is like just a a structural improvement. This is a huge meeting. These meetings have only grown and they will only grow right. So I'm hoping that now that we have like filled out that layer of management that maybe these things will get split up into smaller groups. I don't know what the plan is for that, um but just putting that out there into the world to people who could make that decision.

B

um I think that these meetings have grown too large to be useful and to add the value that they're supposed to add. So hopefully we can split them up, uh but uh that's for other people to solve. For so thank you and then uh nathan.

F

Yeah, so I want to say not to scare anyone, but we have, I guess, one full iteration left um and while we're updating the refresh branch probably every day, um we just need to maybe start qa, and so I'm going to take like 30 minutes a day and then just go through and make sure I don't fix odds and ends, but also remember that, although we're only delivering like a handful of pages for april 27th, like all the pages got affected because of the new nav because of the new styling and so just to make sure that we're covering our bases.

F

So we don't crunch qa into the last 24 hours. But yeah shouldn't be that bad. I think we got it uh lauren. I think you want to add on.

C

uh Yeah I've got that epic there for um like the full refresh branch review, so we can um make sure that our ux teammates get allocated appropriate weight points in our sprint planning.

C

So I think each of them should probably boot up an issue for themselves and then we're going to need when we start doing the final review um like the week of that that project will get really intense too.

A

I added a note there that mobile is not ready on refresh branch. um Most of the components are like mis-sized, so my plan was to at least redo my sections, this upcoming iteration for for mobile, um so be gentle with your feedback on mobile.

A

I think jess is next.

A

uh I just want to call out, because I'm pretty worried about getting images really like illustration assets for or by our due date, before 27 from the brand team. um There's what like two three weeks max left and we have pretty much nothing and we've already tried to get those like devops life cycle.

A

Illustrations we've been like trying to get them for like a year, and we had like an iteration of them from that like two three or four of last year and just to get four of them took like a quarter, and so I don't know how we're gonna get 12, plus the homepage, hero plus, like whatever else we need um in two weeks like I'm. Just I don't know how it's gonna happen. So I'm I don't know I don't know, I don't know what the answer to it is either.

A

um So I guess my question's more like how can we improve this in the future or then? How can we figure this out right now, but yeah and I think hobby you had something about it.

G

Yes, um just plus one, but with content, especially because we're working on templates, like templates, don't work if you don't have content like straight up, they just don't work because there's nothing to there's no variable content to show, um and so I is there a way to invite folks into our planning just to see for them to see like how much of a priority. This is. How can we make the less siloed in the future, not just within this project but for other projects in the future?

G

What solutions can we have to make sure that content is not a bottleneck for our team success.

A

Yeah, I totally agree. um We definitely need to be planning with them um or even have someone from their team embedded on our team so like we can be a priority because they're definitely not a priority for them at all. But we have these due dates that we have to like meet that we need their help with, and so it's like really a bit of a bottleneck.

G

I was typing a follow-up point. I I think we should just come up with our own content. If we're just I'm not, I don't. I don't want to wait till like the last day for us to get all this stuff, and I just like our due date is in three weeks and there's, like you know, for some of our pages. It's like. If we're not getting content like, I think we should just put up something and try our best to iterate towards it and see how that goes.

C

My my last point was directly tied to this desktop screensaver. I have answered your own questions um yeah. I just I don't think we're gonna get it. So what do we do?

C

Not not the answer we want, but or.

H

Images would it be reasonable um to ask gem three for a set of like five.

H

You know, placeholder type, images that say something, but nothing, um maybe not for the home page, but you know for everywhere else, because the stuff we saw today, I'm like oh, we can use. We can use these in a lot of places until we get proper images and then at least we have that and then somebody's not happy with it. We point them to making a brand an issue for the brand team.

A

I don't know how accurate this is, but michael mentioned in one of our one-on-ones kind of as like a throwaway like worst case, we'll just take screenshots of our product and slam them in for for now. So maybe that's a plan, but I don't know how serious is the plan actually. So what the brand team wants to do right now is an nbc one where we take screenshots of our product.

A

But it's not quite that easy because, like we ran into the same issue like two quarters ago, where it's like, okay, like what is like what are we taking a screenshot of and then within that screenshot? What are we highlighting- and I know like matt and cormic- were going back and forth trying to figure that out but, like I said it took them a quarter just to get four of those going.

A

So I'm like how are we gonna do all of this, even if it's literally just taking a screenshot in two weeks plus everything else has happened. That's where I'm like a little bit concerned like no, not in like. I couldn't personally even do that because I don't have access to all the functions of git lab, even if I knew what they were. It's like you have to bring someone else in who has the access you can take the screenshot and, like it's a lot.

A

I'll get going with my next point uh and just keep time um just a quick thing about the release team. I've been working a lot with them back and forth since we moved the homepage over and they need to update their section that announces the new release, um so I've gotten all their documentation updated, but now we're changing the home page, it looks like releases might be living in the resources section at the bottom.

A

I just need to make sure that I'm like linking them to a direct line, that they can change their release number and their link to um and that that stays pretty consistent, um and I know that we're changing things so just you know putting it out there that hopefully we're able to kind of find a consistent place for that to live and it won't change. So we don't have to keep changing the documentation for them. um So it's something to keep an eye out. For um I see nathan typing.

F

Yeah, sorry, just out of something I'm wondering if it's worth it to separate that yaml file, instead of having a giant homepage, one to almost give them their own like, then they can do whatever they want in there and they're not gonna break the whole homepage.

A

Yeah, it's just that the um if the releases live in with all of the resources, like that's an.

I

A

Of cards so that does kind of have to live there, but.

I

You know, I don't know, I don't know, I think.

B

um Probably the I think what might be the most resilient process would be for uh now, because this would like sign them up for more like, like really like. We should read from them like they shouldn't edit us right so like we should have a part of the next build step that, like pulls information from a place like, I guess, here's the thing they can. Actually they can keep their process to some degree. um There is a gitlab api that we can hook into and we can read from files and so like like.

B

If, if we want to make this easy for them, um we can like we should learn their process.

B

We should find out what data gets updated and next at build time, can pull from a gitlab api to get the information it needs, rather than having to get them to like spin up our stuff and learn our new process and then like coordinate with them, like as things change, because they will like let's try and get them to a stable point where they're publishing data that we can trust and that is like well defined because it already is because they've been doing it for years so like since their process is more stable than our process right now, let's rely.

B

Let's have our dependency be on them, not the other way around.

B

And this is the beauty of why this is like. Why? Next is great, because that, because we like get yeah, we can do fancy things with like plug-ins and middleware to like grab this data, and we should have. We should like have some fallbacks.

B

We might even want to like cache it ourselves or something right like like do something you know so that, like we don't break it, if it like api, is down and then like every build while api's down is broken like something like that, but like we should ingest their data rather than having to get them spun up on our new stuff, which is like sort of still nebulous and like. I would be hesitant to tell them how it's gonna be forever and, like their information, is much more stable than ours.

A

Sounds like something we can work on after april 27th, but for this upcoming release like this month, we'll do it manually and then and then going forward. That would be great.

B

Yes, so yeah, sorry that is the like. That is, I think, the goal post. uh It is perhaps not the today solution.

A

Cool uh miguel, I think you're bringing up the last point.

E

um Yeah, it's just like a kind of a self reflection and I found that um I go for the most complicated solution first and I'm violating the the I in credit. So I have to get back on track so I'll, be as an action item I'll be reviewing that section of our handbook and uh yeah.

E

I think lauren.

C

Yeah I just wanna yeah, give you praise for the iteration value diving into it, that one, mr about um adding. You know multiple templates to one url path. It's kind of a little tricky. There's, there's a few different forks. We could take so um it'll be fun to figure out where we go.

E

Yeah, thank you and that's. It.

A

Yeah, I think, we're at time so we'll end the reporting there. Thank you, everyone and we'll see you next time.