 All right. Yeah, I guess we can just start in. Hello, everyone. I'm Andreas Linsen. I work as a designer at Red Hat. Yeah, during my day-to-day work, I work on a product called Compact, which is a server administration UI. But I've also been involved in the GNOME project for about 10 years now. And today, I'm going to talk about one of the GNOME apps called GNOME Maps. And how we use the inter-use to implement a new functionality in that software. So it started when Michael Slumblad, one of the developers, he followed his bug. And he said, yeah, we should probably get some support for public transportation. Because currently, we have that you can get from point A to point B either by walking or biking or driving. But he was like, well, what if someone wants to take the bus? So yeah, we would need a design for this. And I thought, yeah, this should be fairly straightforward. So I grew up in a very, very small village in the countryside of Sweden. It's called Bekemen, about 100 people live there. I didn't grew up there. I grew up up there. So it's quite far away. It takes you two kilometers to get into this place, where the bus goes three times a day, if anyone gets aware. And these days, I live in the central Gothenburg, Sweden's second biggest city. And I have a bus or a tram if I want to go somewhere about every third minute just outside my door. So I felt like, well, this should be easy. I knew both the streams of this, right? I'd be living in a place where it was very, very few buses and trams. And I lived in a place where it's like very a lot of trams. And maybe I took a bus in Paris at some point also. So I should know all this. And I would just need to use my regular process of going from idea, making some basic ideas of who would use this, make mock-ups, and then do the code. And also look at, oh, how is other software solving this problem. That should be easy. But then I did the mistake of talking to my girlfriend about it. And she is born and raised in Rio de Janeiro in Brazil. And we talked about public transportation in general. And it turned out that we had very different experiences and expectations of how one would use public transportation in general. So it seems like I actually had a blank spot here. I mean, I knew how I use public transportation, but I actually don't know how others use public transportation. So I had a problem. So I was like, OK, how do I solve this? How do I address this problem of a great black hole over here? I had read a little bit of literature before about doing user research. But I was always a bit afraid of it. I was always a bit intimidated by it. I felt like, well, if I would do this, I would probably do it wrong, right? Because I wouldn't ask the right questions, and I would just screw up the data, and it would just be terrible. But I thought, well, at this point, I should probably just try it once we're all right. And even if I do it with just a couple of people, I don't have a sample group of 100 or something like that, at least if I get to talk to at least a couple of other people, then that would be better than just basing the data on one person. So I called up some friends and family from different parts of Sweden, and I called some people in Brazil. And I just asked them some very simple questions. And these were all like my friends and family. So I totally understood that, OK, yeah, this is not the best selection of people necessarily, but this is what I have, right? These are the people that I have access to. So I asked them these very simple questions. I mean, the top one, as they talked about in the earlier talk, is a total close questions. But the other ones, I tried to make fairly open. So for example, the top one is like, do you own a car? Which is the question, are you only dependent on public transportation when you need to get somewhere? I intensity of usage, what app is a website to use today, which is a good way of figuring out, OK, what other kind of software can I look at? So I got a lot of tips on different apps people used, what they disliked about those particular applications, and what they liked about those. One was regarding, do you value shorter travel time or less switches, which is a weird question, right? And some people struggled with that. But that was also in order to figure out, like, oh, a sorting algorithm for the different results. And a bit of my walk-in between the stops, also a sorting algorithm thing. And do you prefer a certain need or transportation? Do you like to ride trains more than you like to ride buses, maybe? And also, of course, it worked out to arrive on a departure like when that is important to you. Yeah, so I just call up these people. It was fairly short interviews, maybe like 10, 15 minutes tops. And I was asked these basic questions, and it was quite good, because once you got them talking, they were happy to share a lot of their general experience around these. And they were telling, like, oh, I go from home, and I go, I take these trips every day. And I have these telling us when I'm doing the strips. And with regards to, one of my friends asked, will you prefer a certain need of transportation? And she answered, like, well, I don't prefer, like I said, so much that I simply cannot take buses. It doesn't work for me, because I get really carsick, even on any, like I have to take the trams. And I was like, well, how carsick do you get? And she was like, well, I got so carsick that I just faint. So buses doesn't work for me, right? So suddenly, it occurred to me that this wasn't so much of a preference as like it's such a basic need for her to be able to sort out, yeah, in this case, buses. Because yeah, any kind of results with buses would just be useless for her. It would just clutter up the whole interface. So that was revealing. And I learned a couple of other things as well in a similar vein. So what I did once I had talked to these people, I wrote it all down, like all the results. I put them out on our wiki, the Ignom wiki, and shared it with Marcus again, the developer. And then I tried to summarize the data, and I did this in the form of a person as then. And all of these, as you can see, like they are combinations of data that I got out of the interviews. All right, so here's one of them. This is Jenny, it's a little bit of like just background, which is like, okay, she's new in the city, right? She can't find her way around. And then like a small little story for her and a specific scenario that she needs to carry out, like that a person would need to carry out with this kind of UI. But most importantly, she gets car sticks very easy. So she, yeah, wants to avoid to go by bus. And that I lifted directly from this friend of mine. And the friend that mentioned the car stick mess, she was not the only one that also happened to another person, but she was the one that had had it most severe. All right, and after that was done, and Michael said like, yeah, this is totally fine. Yeah, these things, they make sense. I then went and implemented the UI. And, or I did the wireframes, right? And, yeah, and then I basically used the things that I learned. So it has this simple little family you can sort out, certain means of transportation. For example, if you get very car stick, right? So that then matches the gender persona. You get results here. And at the beginning, if I had done it, like without doing these interviews, I would not have included the numbers of the buses here because I was like, oh, you just go in and shouldn't clutter that too much UI, right? But I figured that a lot of people are actually, they know their city that well, that they know like, oh, I know where I can find tram number eight. And I know how to get between the end station of that to number three, right? I know where I need to switch. And also the leave now is the defaults and it has a dropdown for arrive by, because it made sense as a default. All right, yeah, and then, yeah, I talked again to Michael's. A very important thing I feel in general is that the designer and the developer needs to have a constant collaboration. It can't be that you drop this perfect thing and then you hand it off and then you back away, right? That's not my style. So yeah, he implemented it. And once the implementation was in fairly good shape, I checked out the branch and then I invited some people over to just use some very basic usability testing on this UI and wanted to try if I could carry out these actions. And I did this over like, maybe I did like, person every other day or something, five people in total. And as we found issues, I reported to Michael's and he was able to fix them before the next person would try it out. So the people would have run into the same things over and over again. And yeah, that's it basically. The whole thing wasn't just built by me and Marcus, but it was actually built by all these people who was kind enough to offer the time to do the interviews and also the people who are kind enough to help with the usability testing. So yeah, I guess in the end, we were quite a bunch of people who were involved in building this interface. And yeah, I think that's it for me. If you have feedback about this talk, good or bad, just go to this URL and submit some feedback. And that's it for me. Yeah, so no question. I see what it means. I'm gonna repeat for the camera. So the question is, since my sample has only six people, it could be the issue of that, for example, with car sickness, that I actually happen to run into the one person that has this very weird, because I'm not saying that car sickness is one of them, but it could be that you get very skewed data because the sample is so small, right? Yeah, and I think that is so true. But at the same time, I think it's better if you go out and at least get some data. And at the same time, being aware that, okay, I will talk to six people, could be totally off, but at least I spoke to six people about it, you know? And then hopefully, this is not a done thing, of course. I would love to interview more people. I think also that, yeah, since these things take a bit of time, right? It took me at least 15 minutes for each person. And then I also had to talk to them a little bit more. And so maybe it's 25 minutes in total. It takes a lot of time, right? So ideally, I as a designer on the team should not be the only one that does this, but ideally also other people could help with this. The developers could also be involved in doing the interviews or in doing the usability testing, because then you suddenly, you're able to multiply very easily. Yeah. So during coding monetization, in the sense that Microsoft had done a branch, where the basic functionality was there, not everything, not all the details were in their final state, but it was good enough that it was functional enough that people could actually start using it. And then that's where we started doing the whole thing. And we picked up on very specific things which would have been hard to find otherwise. Like it had certain focus issues, which we had to work out, which is just like, it's such an embarrassment when you sit there and taking notes and you're like, ah, the focus shit, okay. And also in the sense that people use the map itself and not so much the sidebar, I can show here. They would like right click on a map and like do different action there, which was also something I had missed completely. So that's also something we did. But it was good because then, yeah, if we had gone onto a weird path, it would not have been like, you know, 35 coding hours invested in that for nothing. This all is, well, I have at least three minutes so I can give a little demo. Let's see, how do I get the mouse on the right screen? Is it good? All right. So I want to get from UAB to, oh, I'm just gonna butcher it. If I say it wrong, is it connected properly? It's taking a little bit of searching unfortunately. All right, so from here and then, I'll say this there, who belongs? Wow, what's up with that? It is not in master yet. UB, all right? U-L-B, and yeah, here we get the results while it's broken, I'm so sorry. It is, it is, I should have, I should have recorded the whole thing, right? Okay, here you have the same bar, you get the difference of results. Okay, if I select to go with numbers 25, oh yeah, I get the stations in between, which is an expandable load of thing, which is good if I'm new in the city because I need to, oh, what is the station before the station I need to get off, right? And I also get walking instructions for how to get from the subway to the street and then the arrival time. And then I can go back, I can choose the different. To sort out ferries, they wouldn't go in a ferries there, but in some cities, yeah, so that's it. I can also load later alternatives, okay? I can go at 16.45, and yeah, that's how it works. Thank you for sure, sir. Thank you. From a project called, so the question was where did I get the data? There is a project called Open Trip Planner, which is an open source project. And there is a couple of cities that pass their data in this kind of formats. We currently have a test server up and running. We need a proper server, which we would host ourselves. We haven't solved that yet. It's all good, I think. And I'm out of time.