GitLab GitLab 13.5 Release Kick Off, 17 Sep 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Create:Ecosystem 13.5 Kickoff

Description

Issues discussed:

- Display Repository information in Jira: https://gitlab.com/gitlab-org/gitlab/-/issues/229349
- Add data to Atlassian Connect application: https://gitlab.com/groups/gitlab-org/-/epics/3915
- Add Issue Detail view from Jira Issue List: https://gitlab.com/gitlab-org/gitlab/-/issues/220999
- Drop 'default' column from services table: https://gitlab.com/gitlab-org/gitlab/-/issues/217581
- Add telemetry for group-level integrations: https://gitlab.com/gitlab-org/gitlab/-/issues/233940
- Add MAU data for Ecosystem Usage Ping: https://gitlab.com/gitlab-org/gitlab/-/issues/225678
- Increase frontend performance of Project Home page: https://gitlab.com/gitlab-org/gitlab/-/issues/233716
- (Dashboard - Internal Only) FY21Q3 IACV: Make GitLab development Pajamas first: https://app.periscopedata.com/app/gitlab/718783/FY21Q3-IACV:-Make-GitLab-development-Pajamas-first

A

Hi, I'm patrick dooley, I'm the senior product manager for git labs, ecosystem group, and this is our 13-5 kickoff. Let's talk about the work that we're going to be doing over the next month, so first up we're going to be adding something to the jira development panel, so we're looking to add repository information and show you where it's going to go. This is jira. Obviously here's some issues that we've put in a board.

A

If we open up this board and work has been done inside of gitlab, it shows up here in the development panel, so we see, there's been one commit and there are two open pull requests which come to think of it doesn't make any sense. Does it, but uh we've got the this command. We have a listing of that commit. We have the date it was made. We have the files that were changed.

A

We also have these two pull requests which are merge requests, but in the development panel they're listed as pull requests, uh it shows the title of it the status of it when it was last updated. So all of this is really cool. It's really helpful for specifically for developers who are working inside of this. It shows them all of the work. That's been happening and allows you to easily navigate around, but it's missing a few things. So we don't have any branch data we don't have build deployment or feature flag data.

A

All of that stuff would be really cool. We just don't ship it over yet so today, all we ship over is that commit and merge request data. So in 13.5 we're looking at repository information, repository information, we'll fill in this branch data, so we'll show what branches are currently active and then in here's, my epic for this in 13 6 we're looking at adding build 13 7 deployment and 13a feature flag. So what we're hoping here is in four iterations we'll be able to fill in all of these tabs.

A

We'll have a much richer set of data there for people to interact with. Also in that jira space, we recently released the jira issue list. So this shows you a list of jira issues inside of git lab the idea being that we would like to get to a position where, if you're a developer, who's working in gitlab, but your project management is happening inside of jira we'd like you to be able to do 80 plus percent of your work inside of gitlab. So this is kind of the second step.

A

First, we showed you a list of issues and now we've just recently or are about to release. As of when I'm recording this the ability to federate your get your atlassian account inside of gitlab and with those individual user credentials, I can now grab details from each issue, so we're going to add a view that allows you to see the details of each issue and then past this we're also going to look at adding interactivity there, so the ability to then comment or change the status or add or remove labels.

A

So this is kind of step, one on those details.

A

So on mass integration, which we've we just recently released instance level integration, we're finishing up group level integration. Now I don't think group level integration is going to ship in 13.5. We have a couple things we'll talk about here, but then there's some more work that we're doing to finish that out. One of those pieces is some cleanup on our database tables. We need to drop the default column from the services table.

A

That's part of cleaning up for the group integration that's coming soon, and then we also need to add telemetry for group level integrations so that'll specifically track right now. What we track is the usage of how many projects are integrated with service x. We also need to track how many projects are integrated with service x via the group level right.

A

So, if you integrate on one group, but there are a thousand projects behind it, showing that it's only integrated with one group that doesn't really tell the whole story, so we need to be able to see the difference there.

A

Also, on the telemetry front, we want to add monthly, active users for the stats that we collect for ecosystem. So again, what we do is today is we collect how many projects are integrated with how many services? What we also would like to know is how many users are inside of those projects.

A

So if you have an integration that has lots of projects integrated but they're, all very small projects that only have one user each right, like that's, not as interesting, necessarily as an integration that has only a few integrations, but there are thousands and thousands of users each so helping us understand. The volume of total users who are using these integrations will give us a much better lens into how we prioritize and think about what we work on next.

A

So we've got two okr things to talk about. One is increasing the front end performance of our project homepage, so the project home page is specifically this page see how it takes a second to load, a lot of things kind of move around. That is the performance that we're talking about and specifically we're. Looking at the largest contentful paint is what we're hoping to address so today, it's 6.6 seconds.

A

We want to get that down to 4.62 seconds and that's all of these individual pieces, as they get painted in the largest contentful paint, is the one where you sing the most things kind of snap. In now. There are a number of ways we can do this. We can reduce the css load. We can reduce unnecessary javascript.

A

There are also, for example, some unnecessary components on every page load that we may be able to remove. So here's a good example.

A

So if you click on the notifications settings and you click on custom, this modal- all of these things- all of this is being composed on page load every time and uh despite the fact that this is probably not very commonly used right like how often are you actually changing your notification settings so things like that and again, css and javascript load reduction can help us drive that number down and then also on the okr front. We've been working towards making gitlab development pajamas first pajamas, being our design system and front-end toolkit.

A

So we've got this issue where we're tracking eight individual components that we're trying to migrate from our legacy tools to pajamas- and uh this is our progress so far- we're 42 complete. So far, a couple we've got two months in one month to go, uh and here's where we're at we've been focusing on view buttons on pagination, we started with 213 components. We have 86 as of today, so we've made a ton of progress on buttons and pagination.

A

We still have a bunch of drop downs, modals tabs pop overs and tool tips to go, but this has been huge, absolutely huge, that we've made this much progress. Every single one of those components is kind of one less piece of technical debt that we're carrying around makes things more. Maintainable makes things more performant to the previous point, right, more performant, easier to maintain and more consistent throughout the application. So this is really important work and thank you to everyone who has been pitching in and participating.

A

This has been really exciting, so that's it for 13.5 tons of awesome, stuff working on some cool things in jira, getting some more data so that we can better serve you and then working to improve performance. It's going to be the best release ever and we'll talk to you again in a month.