 Hello, everyone. We've just approached the hour, so I'd like to show you what UX research we've been working on at the moment. So in my last update, you may recall that I just discussed that the UX team had started working on improving GitLibs navigation. To recap, we tested with six users, and these were a mixture of existing users and non-users. The non-users were typically using a competitor tool such as GitHub or Bitbucket, and we gave each user the same set of tasks to work through. And these tasks were designed to purposely move the user between the sidebar, their profile and contextual navigation, which appears when you're inside of a project or a group. And on this slide, you can see how the navigation looked during our first round of testing. As you can see, it's not too dissimilar from our existing navigation. The main difference is the pinnable sidebar still existed. This design highlighted a massive 30 usability issues. Now, all usability testing sessions are recorded, so if you head over to issue three within the UX research project, you can watch users as they try to get to grips with GitLibs navigation. Since that testing took place, we've gone through two further rounds of usability testing on the navigation. I'm going to spend today explaining in brief some of the changes we've made and how users responded to them. So this slide shows how the navigation looked during our second round of testing. This was a prototype created in InVision by Pedro. It included a link to the prototype in the slide, so if you'd like to explore it, then feel free to do so. The design highlighted 29 usability issues, so we've improved ever so slightly. So let's first talk about a couple of the things that the design did improve. So in the first round of usability testing, we found that it wasn't immediately obvious to users that there were two levels of project navigation. And we found this out by asking users to navigate to the issues board. And a lot of people didn't realize that you had to click issues first in order to trigger the secondary level of navigation and get the board view. So in this round of testing, we changed the navigation so that the items at a secondary level would appear on Harbor rather than the click of primary level items. And subsequently the success rate of this task improved. We also witnessed in the first round of usability testing that users struggled to identify their personal projects from their projects list. And by adding the tab of personal projects on this page, every user in the second round of testing could confidently state what their personal projects were. And finally, in the first round of testing, users didn't realize that the Tanooki is interactive. And this may not sound like a massive deal, but it is for two reasons. And the first is that users were forced to manually adjust the URL in their address bar. And that's a really messy way to navigate. And it's subject to users making mistakes. And secondly, the UX team is planning to create a more personalized default dashboard. And we know that something card users want because they're told us in various research studies and the Tanooki could be one of those ways in which users access this dashboard. So they do need to realize it's interactive. So in this round of testing, the position of the Tanooki has moved from the center of the screen to the left hand side of the screen next to the hamburger menu. And this private subtle change made users realize that they could navigate using the Tanooki. But even though the UX team has alleviated these couple of pain points, you might still be wondering why there isn't a greater reduction in the number of usability issues. Well, that's because when you add new features into a design, such as the new icons for adding a project and assigned issues and assigned merge requests, you can create new usability issues while you're attempting to resolve the existing usability issues. So for our further round of testing, we decided to test two concepts. And again, these are clickable prototypes that users could navigate around. And they were created in-framer by Chris and Dimitri. The main differences between these two prototypes is that the breadcrumbs were displayed slightly differently. So we have a horizontal and a vertical version, as you can see, I click back and forth. Also, the content within the top section of the screen, such as the search bar, to-dos, and so on, were arranged slightly differently. So part of the testing was to find out which set of breadcrumbs would help users to identify where they were within GitLab. And the other part of the testing was to see if it would improve the way that they navigate around GitLab. So both of the prototypes had the exact same changes to the global and sci-fi navigation. So there's been one very consistent pain point through each round of usability testing we've done so far. And that was that users had no concept of the different types of navigation that GitLab has. And we witnessed this in a couple of different ways. But again, due to the limitations of time, I'm going to very quickly talk about the two main ways. So at the beginning of the previous testing, we had a task which took users directly into a project. And it exposed them to the contextual navigation for the first time. And then later on in the testing, we had a couple of more tasks which forced users to revisit the contextual navigation. However, the success rate of the later tasks was really poor. So users hadn't remembered that the project navigation existed. They never mentioned it or actually looked for it again. So this tells us that not only do they not remember it, but they have no concept of it whatsoever. And there are also two extremes of user behavior when interacting with the sidebar. So some users didn't instinctively interact with the sidebar because it isn't open by default. So for example, it took one user around nine minutes to find the sidebar. And those that did find the sidebar easily typically proceeded to treat it as their primary form of navigation. And subsequently this meant that they often failed tasks where the content couldn't be found within the sidebar. So these new prototypes in the third round of testing tried to address those issues. The contextual navigation has now moved to the sidebar. And we now have a global dropdown menu which contains some shortcuts such as groups of projects you've recently viewed. As we had two prototypes, we've doubled the amount of users we tested with. I spoke to 12 users in total and that comprised of eight GitLab users, four non-users, and they were split evenly, so six users to each prototype. So the results, 14 usability issues were discovered, which considering the amount of changes we've implemented since the previous prototype, that's actually pretty good. We've halved the number of problems. Generally the majority of users seem to understand that the sidebar was contextual after six or seven tasks, which is about halfway through the testing. And this is an improvement from the previous research which showed that users could complete a whole usability testing session without understanding that the sidebar was contextual. The breadcrumbs were inconclusive. Half of the users we tested with utilized the breadcrumbs to understand where they were located within GitLab. However, it was uneven split across both prototypes, so three out of six on each prototype used them, so there was really no clear winner there. And we had a lot of positive feedback about the new global dropdown menu and users' behavior reflected that as well. So what's next? Well, we'd like to get the number of usability issues even lower, and you can read what those issues are in issue seven of the UX research project. We've realized from the testing that there will be users who are apprehensive about moving to a contextual sidebar. So one user discussed how by referencing stuff in the center of the screen, he didn't always spot the contents of the sidebar changing as his attention was focused elsewhere. So perhaps there's still some improvements that can be made visually to help users shift their gauge to the new sidebar. And we're also asking some seasoned users of GitLab. I mean, they do tend to sometimes have a better understanding of the existing contextual navigation. And then we're kind of asking them to change a behavioral pattern that they might have passed for a couple of years, and that takes some getting used to. So we want to make sure that what we're proposing to implement works for both existing and new users to GitLab. And we're looking into ways in which we can effectively transition users to the new navigation and keep their frustrations to a minimum. So that concludes my update for today. Does anyone have any questions? Going once, going twice. Now what? Super. Well, enjoy the rest of your day, everyone. Thanks. Bye-bye.