1 00:00:02,000 --> 00:00:02,000 SH Liam 2 00:00:07,000 --> 00:00:47,000 Design Notes is a show about creative work and what it teaches us. I'm your host, Liam Spradlin. Each episode I talk with people from unique creative fields to discover what inspires and unites us in our practice. This season, I'm starting with a special series that celebrates 10 years since the launch of material design, exploring the inception, evolution and future of Google's design system. In this episode, I sat down with software engineering manager and musical theater writer. Will Larche in the interview will explain to me what makes Software engineering a study of choices where the creativity lies in encoding design, and how he imagines that AI experiments could open up an abstract new future for the interface. Liam 3 00:00:48,000 --> 00:00:53,000 Let's get started. Will welcome to Design Notes. Will 4 00:00:53,000 --> 00:01:05,000 Thank y ou, Liam. I'm so excited to be here. Finally. I mean, you took your sweet time making me a guest, didn't you? But I appreciate that you wanted it to be the biggest hit that it possibly could be before you even approached my people. And that's kind Liam 5 00:01:05,000 --> 00:01:19,000 Right. This is long overdue, and this is part of the reason that I'm loving the M 10 campaign so far, because it does give me an excuse to reach out to Will's people and finally get some of my favorite coworkers on the show. Will 6 00:01:19,000 --> 00:01:25,000 I will admit there are some people that are wondering what happened to M four through nine, but we'll explain. We'll make it, Liam 7 00:01:25,000 --> 00:01:36,000 We'll make, yeah. Don't worry about it. . Uh, okay. Will, for the listeners who don't already know, who are you and what are you working on? Will 8 00:01:37,000 --> 00:01:50,000 You know, I ask myself that every day. Liam . Uh, I am Will Larche, I am an engineering manager here at Google. I am, uh, now the lead in Google design platform for AI for designers and developers. Liam 9 00:01:51,000 --> 00:01:56,000 Okay. As soon as you say that, there are gonna be people wondering what that means. Will 10 00:01:57,000 --> 00:02:19,000 So I manage global cross-functional teams that attract and retain users to the Google ecosystem through improving its look and feel and save the company money by increasing their productivity efficiency and creativity of designers and developers through the application of large language models to common problems. Liam 11 00:02:20,000 --> 00:02:26,000 I wanna know, first of all about the journey that led you there from the beginning. How, Will 12 00:02:26,000 --> 00:02:29,000 What, what beginning birth, where are we going? Liam 13 00:02:29,000 --> 00:02:32,000 You should start where you see the start. Will 14 00:02:32,000 --> 00:03:12,000 It started for me when I joined a Buddhist meditation group in the West Village, um, called Friends of the Western Buddhist Order Now called triratna. And, uh, their leader did it for a Georgian economic publisher. If you're not familiar with Georgian Economic Theory, I'm not here to espouse it or, or, or, you know, promote it. But it's a, it's a, it's a theory of taxes based on property owning instead of income. And he was going on vacation and knew that I was kind of like computer literate. Like growing up in the nineties. Will 15 00:03:12,000 --> 00:03:39,000 You had to learn how to take care of a computer. You couldn't just have one that worked. That was not, that was not the goal of the companies that were making computers back then. It was more like a fun hobby that you were always building your computer. And he said, can you come in twice a week and restart the printer for the bookkeeper? I was like, yeah, I can do that. And so then, uh, I did that and he sent me this message while he was away saying, we need to change the copy on the website for the charity. Will 16 00:03:39,000 --> 00:04:20,000 And I was like, oh, okay. I'll try that. And I went into Dreamweaver. If everybody remembers Dreamweaver, um, I believe you can get me through the night. Um, it was a simple task. Change a sentence on the website. I could not do it. I could not figure out how to do it. I did a poor job. It made me mad. I went home, I was super upset, and I said, maybe I'm blaming Dream Weaver. Maybe the real person to blame is Will, because Will didn't know anything about this. And I said, if I knew about this, if I understood what I was doing, would I like this? Will 17 00:04:21,000 --> 00:04:53,000 And the answer was yes. I was pretty sure that I really would like working with, you know, website as I thought at the time. And so then I decided to get some education. I ended up taking a class at, uh, NYU Night School called Intro to See, which was like, what's a compiler and how do you use the plus and minus operators? Uh, and from there I talked my way into an internship telling them I wanted to be an iOS developer. 'cause iPhones were new at the time, and I was super excited about how nice they were and how beautiful they were. Will 18 00:04:53,000 --> 00:05:21,000 I wanted to make apps and I wanted to understand it. I got that internship, it turned into a job. I, uh, then got another job. I rose at that startup to head of mobile development. Um, then I ended up becoming a VP of engineering somewhere, transferred around to a lot of other startups and agencies. I was chief project officer and mobile lead, that one at the same time that Google was reaching out to me. And I said, no, Google, I'm not interested. I don't know why you'd think I'd wanna work for you. Will 19 00:05:22,000 --> 00:05:54,000 I had no interest in being a cog in a machine. But then my startup closed and I said, Hey, Google, let's talk. And then I actually turned down several teams that, that were interested in, uh, having me work on them because they just didn't feel right. And then they said, well, have you heard of material design? I said, heard of Material Design. When I was managing designers back at that startup, I didn't really know what I was doing, but I had one that created her first mobile designs ever. And they didn't really work, they didn't make any sense. Will 20 00:05:55,000 --> 00:06:31,000 And I had seen material design come about these external guidelines. And I said to her, look at these and see if they could help give you some knowledge or inspiration. She did two days later, gorgeous mobile designs that were usable and buildable that turned into our product. So I had already seen the power of material design to help someone like fundamentally and change essentially the, the path of a product from being uhoh to Wow. So yeah, I was excited to work on a material design. I came in as an iOS engineer and mobile and motion expert. Will 21 00:06:31,000 --> 00:07:02,000 And then I ended up founding our Flutter team that ran that for years. I ended up running our color team and our color space, the HCT color space, material color utilities, uh, our work in personalization. And then, um, my engineering director asked me to start, uh, an AI initiative. And that's turned into a bit of a thing. I've also built Figma plugin tooling. I think I sent you a list of all the things that I've worked on. It's actually a lot, I think for, for for one person. I bounced around, but always within design engineering here at Google, Liam 22 00:07:03,000 --> 00:07:17,000 To be honest, I didn't, I didn't even need the list. I feel like the, the stuff that that you've done has had such a big impact on the team and by extension, like all the people building with material that it's like impossible not to know the list that is Will 23 00:07:17,000 --> 00:07:17,000 So kind. Liam 24 00:07:18,000 --> 00:07:57,000 So something between flutter, the HCT color space, working on Figma tooling, um, also this perspective that you have about making things more efficient for the people who are making software. Something that I, I really want to ask about is the creative nature of making software from an engineering perspective, because your work intersects so often with design, and I'm curious like what your perspective is on the relationship between those two things. Will 25 00:07:58,000 --> 00:08:33,000 It's really weird. So engineering is the study of choices. We have the thing we need to build, and there's 10 to a hundred different ways to do it. And some of them are obvious, some of them are more avant garde. You know, Hey, you wanna go ride everything in, you know, machine code fine. Or you can use the newest framework that hasn't been battle tested yet, but you have a good feeling about those are the choices that engineers have to make every day. So there's something creative to that. And then once we've made our choices, we sit down and we write. Will 26 00:08:34,000 --> 00:09:12,000 So it's very much like being a novelist or some other kind of writer, except that there's a right answer in the end that you can write whatever you want, but it is always supposed to do something that's been predetermined. We've gotten requirements, we have a spec, we have designs, maybe if we're doing, doing front end work, and it's supposed to look like that and go like that. So there's all this creativity towards a constraint, which I find very different from visual design. When I watch you all work, I've often seen you all do projects where it's like, Hey, what's next? Will 27 00:09:13,000 --> 00:09:42,000 Where it's literally just like, we should be thinking about the future and, you know, what does that mean? And then, you know, Karen ing, one of my favorite designers goes, I'm really interested in mycelium the roots of mushrooms right now. And people go, wow. And then that turns into sparkles and material three and stuff like that. Like, it's a a totally weird process where you all don't have a right answer. You have to just create and do stuff. And it's very open-ended. And that's, that's a little intimidating to me. Will 28 00:09:42,000 --> 00:10:01,000 That's why I I consider myself not at all a designer in any way, but a design fan. I am here because I love what you all do so much, and I wanna make it come true for the user, and I wanna give them a high quality experience because of the amazing ideas that you all have, the, the magic that you pull out of the air. Liam 29 00:10:03,000 --> 00:10:50,000 Okay. A couple interesting things there. First, Karen is absolutely right. Everyone should be paying more attention to mycelium . Yes. And second, it's interesting to hear your perspective on that because I always think about, you know, when I talk to people of, of all kinds of different creative disciplines, I'm thinking about like, what is the material that they have to work with? How do they manipulate it? And what are the constraints that are placed on it? And I think for me, even talking about some of this like visioning work or like discovering a new vibe or, you know, coming to some conclusion about how the interface should work in the future, even there, like, it seems a little bit more clear to me, but I I'm really interested in this comparison you made between writing code and writing a novel. Liam 30 00:10:51,000 --> 00:11:02,000 Like how, how do the materials and the constraints work together when you're, you've made your choices about how you're going to approach it, and now you're writing, and then what happens? Will 31 00:11:03,000 --> 00:11:35,000 Well, you are making the assumption that we've made our choices. And, and I think most of the time that's true. Sometimes there's just pure exploratory software engineering, but oftentimes it's about the fun of it. So I guess, you know, knowing what I know about all the arts, you are manipulating whatever you have either until it does the thing that it's supposed to do, or it gives the feeling, which I assume is what you're probably doing in the visual arts, right? Is that you just have this intuition and feeling that it's done right. Will 32 00:11:35,000 --> 00:12:06,000 Yeah. 'cause that's what happens in my music and in my theater work too, is I, which I haven't mentioned yet, but this is all, um, a part of a big ruse for me to be able to do, um, queer and avantgarde music theater, um, in the East Village. And that's very expensive. And so thank you Google for supporting me. Um, it's, it's difficult sometimes because it can seem like a sea of I don't know what to do and it could be anything. Um, but, and then I've also run into where I've made stuff and people are like, why don't you just keep it? Will 33 00:12:06,000 --> 00:12:45,000 That's good enough. And I'm like, I don't know how to tell you that it's wrong still. This still isn't the right thing and I need to tear it down and build it again. And it can be very frustrating for other people to see a, a writer do that. Um, so engineering has an appeal to me in the sense that I live in this world in the arts that's total ambiguity and is so difficult. And then I can come into work and there is a right answer, but still requires some creativity on our side. I think as time goes by in the industry and there's more constraints based off of resourcing and, you know, uh, we're not just in hyper growth mode anymore. Will 34 00:12:46,000 --> 00:13:15,000 We do actually expect people to make more choices ahead of time and to go ahead and build the thing, quote unquote the right way the first time. But I'm smart enough as a manager to know that there is no such thing as the right way. We need to revisit things. We need to do postmortems on work all the time. We need to refactor, we need to build in time to change things. Um, and then align that to when business priority shifts so that we are making the next version of the thing when it can also have a difference for the user. Will 35 00:13:15,000 --> 00:13:25,000 And so as a manager, I'm often trying to enable my people to be able to do their creative work, to learn from it and to improve things while also intersecting that with what the business needs. Liam 36 00:13:26,000 --> 00:13:45,000 Yeah, there's an interesting recognition there that like what is the right thing is so contingent on the rest of the context around it. In engineering, in design, certainly in writing, if I am writing something and I'm working on a draft over multiple days and then I come back, I can definitely tell what mood I was in when I was last working on it. Will 37 00:13:46,000 --> 00:14:30,000 There's also this phenomenon in engineering where every engineer thinks the other engineer is wrong. It's very weird. And I think that's, uh, a function of our human brains where if one engineer is working on something large, they'll spend so much time putting themselves into it. And like I said, dealing with thousands of choices, every single thing that they decide to write, the way that they name a variable to whether or not they're using a nip statement or a switch. You know, sometimes these things are predetermined, but most of the time it's up to the engineer to decide These are the very personal pieces of engineering that then when somebody else comes in, it's like when you're in somebody else's house, and you might most of the time feel like, oh, I love this house, but I would do it different. Will 38 00:14:30,000 --> 00:15:05,000 You know, like, yeah, like not that rug, you know, that sort of thing. Engineers have the same problem where even if they like something, they're like, well, I would've done it differently. And I think one of the mistakes engineers make is thinking that that feeling is truth that has to be followed all the time. Instead of finding a way to learn more from other people. The best engineers are the ones that are willing to, um, look at what's going on, learn from it, and then figure out what they would do next, not what, what they would've just done in the past. Will 39 00:15:06,000 --> 00:15:11,000 I mean, that's useful in, in a postmortem or something. Um, but most of the time we're trying to move forward. Liam 40 00:15:12,000 --> 00:15:46,000 Yeah. It's, it's interesting to make the comparison to like how someone has decorated their house because at least, um, for myself, I certainly think of code as something more objective than design. So it's interesting to hear like how big a part of it actually is, like the subjective input of the person writing it, and all these small decisions that I feel like in design, I really have to be conscious about keeping up with that so that at the end I can understand and also convey that context to someone else in a way that it makes sense. Will 41 00:15:47,000 --> 00:16:20,000 It's definitely underappreciated how much is, um, the million little choices inside the code and how the person who was writing it felt that it was beautiful or still hated what they were working on. Um, I don't know if designers run into this, but I think most people when they're working on things can be like, Ugh, I hate this, but I have to move forward. Sometimes that's actually when you do your best work though, is when you have to accept, um, that you couldn't get something to totally please you or to make it perfect, but then you're able to ship and have an impact. Liam 42 00:16:20,000 --> 00:16:23,000 Yeah. Disappointment can be focusing. Will 43 00:16:24,000 --> 00:16:30,000 Uh, yeah. And you know what? Disappointment's played a very big role in my life, I'll tell you that much. Liam 44 00:16:30,000 --> 00:16:31,000 Do you wanna talk about that Will 45 00:16:32,000 --> 00:16:35,000 ? No, I don't, Liam. That's very personal. Why would you even bring it up? Liam 46 00:16:36,000 --> 00:16:37,000 I and on that Will 47 00:16:37,000 --> 00:16:56,000 Do, on, on the, on the house decorating thing though, you've been to my house. Mm-Hmm. , you know, you know how I go. I'm always aiming for, since I'm such a fan of design, but I don't have any talent for it. So I hire really great designers and tell them to go nuts. And I, the kind of feeling that I want to someone walking into my house and going, oh wow, that's a little much Liam 48 00:16:57,000 --> 00:16:58,000 , that's great. Will 49 00:16:58,000 --> 00:17:02,000 That's me, right? , I'm glad my husband feels the same way. Liam 50 00:17:02,000 --> 00:17:27,000 It's a, it's a good spot to aim for. I want to go back, I don't wanna brush under the rug all the stuff that you were saying about your work in musical theater and writing. I've been in the audience at some of your shows. Um, I wanna talk a little bit about that, and I also wanna talk about how, like, the interplay between that and the work that you do day to day. What, what does that feel like for you? Will 51 00:17:28,000 --> 00:18:09,000 Well, it feels like I'm acting like an engineer. I, I deal with my, my psychoanalyst, he always tries to tell me like, do you feel like you're performing? Yes. Yes, I do all the time. Doesn't mean I'm just ingenuous. It just means that I do feel like there's a heightened sense of being aware of how others are perceiving you. That is very valuable if you can wield it. And I think one of the reasons why I've had exactly the kind of trajectory career that I've had is because I'm willing to bring personality and a curiosity about people and empathy and essentially, you know, my performative skills to something that can be kind of dry. Will 52 00:18:09,000 --> 00:18:47,000 You know, software engineering can be a little bit like, Hey, just leave them alone. Let them go. Stay in their room, don't bother them. What I really, really want is for people to understand each other and to always come at whatever's going on with a maybe attitude. Whenever designers, product people and marketers, whatever, come to an engineer, the engineer needs to say, maybe listen to it, evaluate it, and give them the time to see that. It might still be no, but like, don't start with No, that's, that's such a, a way for good ideas to get lost and for, um, amazing things to never get built, unfortunately. Will 53 00:18:47,000 --> 00:19:11,000 Yeah. But it is, it is like kind of like this tradition in software engineering that we are crusty and that we put up a front that tells people, um, leave me alone. Mm-Hmm. , we actually don't do our best work that way. We do our most self-indulgent work that way, because then we are just writing for ourselves. Yeah. But there's, there we are collaborators, there's clients and there's end users, and all those people are the responsibility of the engineer as well. Liam 54 00:19:12,000 --> 00:19:25,000 Yeah. And I, I think design and engineering have shared goals around those audiences. And I also think like we are all in one of the best positions we can be in to make these things possible. I think, um, Will 55 00:19:25,000 --> 00:19:25,000 What do you mean? Liam 56 00:19:27,000 --> 00:19:44,000 Meaning that, you know, we are in a position at Google working on material design lots more resources than in, in, you know, certainly any of my previous jobs, uh, to find the solution. I think if we can't find it here, I don't know if it can be found , Will 57 00:19:45,000 --> 00:20:29,000 I'm gonna just agree with you. I think we have far more constraints these days than we appreciate. One of the things that's been on my mind a lot lately is why is that? And how can we fix it? Because we do have this responsibility to make sure that with 200,000 people working for us, we are doing great work in moving fast. And we have this way that we've worked for the past 20 years of building shared infrastructure, like material components and like the many, many things that we have internally that run our backends and our front ends and , I, I would just like the people on the outside to know that like engineering inside Google is completely different than it is anywhere else because we have custom shared infrastructure. Will 58 00:20:30,000 --> 00:20:56,000 And every time somebody builds something like that, any little thing that we're building that's supposed to be shared, we're making this promise that we're going to help the teams that are using it move faster. But I think it's actually worth asking the question as to whether or not they do. If we want to change something, we end up breaking somebody else's application that's doing it, or we have to talk to a hundred different teams and sell them on what we want to do. And what if only some of them do, and then we have to align with more people. Will 59 00:20:56,000 --> 00:21:26,000 It becomes harder and harder to move forward every day. There's also this tension in material design between are we building blocks or are we creative direction? And we've kind of always agreed that we're both, or somebody actually, when we started, when right before I started here, like when earlier teams were releasing the first material, there wasn't components at all. It was no building blocks, it was just creative direction. And we changed the world that way. But then people kept saying like, well, you want me to have a button that looks like this? Will 60 00:21:26,000 --> 00:21:54,000 Give it to me. And it seemed reasonable. And my team, when I started on the iOS team here, it was actually people inside the company building a button, being like, Hey, I am on maps and I built this button. Would you like to use it on search? And then somebody being like, yeah, put that, put that somewhere where I can copy it. And then that turned into a directory of material components that people had made. And then material was like, should we help with that? And people were like, yeah, it seems reasonable that we should help with that. Will 61 00:21:54,000 --> 00:22:22,000 And then over time it became creative direction can't change without the components being ready. And we released these things together and people keep asking for that. That's what they expect. But it might be that that is still missing the mark in some way in being able to, um, bring true innovation. And I wonder what it would look like if we had a separate team for creative direction and a separate team for building blocks. And the building blocks were really, really good at being flexible. Mm-Hmm. Will 62 00:22:22,000 --> 00:23:03,000 . And they could build custom UI much more easily because secret, even in a, like a material app, only 40% of the UI is like material. And then the rest is custom to the content of that application. And so I think that in a, in a better world, we would separate these things and if we had components, they would be so flexible and divorced from their styling that we could change them at any time. So we, we really should be building things in a way that they're so flexible that we are indemnified for a future we cannot see instead of, um, calcifying the spec of today into components that are difficult to change. Liam 63 00:23:04,000 --> 00:23:48,000 Yeah. And, and I do wanna talk more explicitly about yeah. Where material was when you joined and where it is now and how we got there. But I do think there's a really good point there that, uh, we kind of progressed from, or emerged from a sort of collaborative, like fast moving, chaotic way of assembling designs into something highly systematized, in which like our concept and perhaps others' concept of how to implement a design system was along these very objective lines of like, components that look exactly right. Liam 64 00:23:49,000 --> 00:24:24,000 The shadow values are exactly how they should be, the size, the typography, everything. Great. Um, and that now we're kind of moving with relational color and all of those things into a space where we're trying to find a way within those boundaries to, as you said, like communicate the creative direction or communicate more of a mood, more of the subjective quality of the design and, and potentially facing a future where we have to build those flexible components that maybe we should have built from the beginning. Liam 65 00:24:24,000 --> 00:24:33,000 But I wanna talk a little bit more about kind of how you've seen that progression from where you started in concrete terms, and then we can extend into the future. Will 66 00:24:34,000 --> 00:25:13,000 Well, I think the, the biggest change was it used to be designers were just making guidance and then everything else hopefully happened. So this was kind of this upside down world where design was completely in charge. They were still, you know, not having a good collaboration with engineering as is so often the case, but in a, when the design system is the product they could just go and do. And I think that's kind of rare. Like, I can't think of many other products where the engineering is second maybe or not the most important thing. Will 67 00:25:14,000 --> 00:25:45,000 The components were not always seen as necessary. It was our clients that told us the, they were necessary because it wasn't going to come true for most of them unless we supplied components. 'cause that became the pattern. And it seemed like I said this sort of reasonable expectation that we should always have components that match the guidelines. So eventually we got to this point where they would tell us that the guidelines were changing or had changed. And then we got to this point where we said, Hey, stop releasing guidelines until the engineering is ready. Will 68 00:25:46,000 --> 00:26:23,000 And then these things kind of became, they marched in lockstep. Um, again, what sort of happened under the hood that was more difficult is the engineering is then the blocker. Not for being built, but for being shipped. Because we have, you know, hundreds of clients with inside our own company, much less on the outside. And every time you change something, people complain. They've maybe customized it in some way and you didn't know it. And then they're like, Hey, I was using this button as a date picker and now it's not working. Will 69 00:26:23,000 --> 00:26:52,000 And you'd be like, oh, well, at Google our, we have this idea of a mono repo, which is one place where our code is stored, and it would be our responsibility to resolve that problem, even though somebody was kind of using off script behavior. And it would, same thing would happen on the outside where people would be like, give us the next material. We're so desperate for us. Why are you holding it back? Why are you so slow? But then if we did it in a way that we were changing what they had an even larger contingent of, people would come forward and say, you just broke my app. Will 70 00:26:52,000 --> 00:27:24,000 I did not ask for material three. My app doesn't even look like material two. But I built it off of material two to do this other weird stuff. And since we had done it in this way, that wasn't really built for flexibility or flexibility was kind of always something that was just kind of figured out locally that then it, it became impossible to move forward that way. And I think that tension between are we building blocks or are we creative direction? That's, that's the sort of largest, the largest conflict that I see still going on inside material design. Will 71 00:27:25,000 --> 00:27:59,000 Some people think we still don't have enough people to really even do what we're trying to do, which is to power the entire expression of Google brand through software, as well as enable people in Android to do the same thing for all of their brands, or to use a built-in baseline that they can theme on top of that and addition. So on top of that creative direction to also just be how you build apps to like standardize that we are the button, we are the text field and that sort of thing. And each of those components is a tiny application. Will 72 00:27:59,000 --> 00:28:24,000 That's huge. Some of them are huge. I worked on text field on iOS, it's, I, I once printed it out because somebody was like, I don't think you're technical enough to get to the promotion you want. And I was like, hold my beer. And then I went and printed out all the code that I had written for Text Field, had it bound and dropped it in his lap and said, I've written this much code. Did you know that? And they didn't because they thought it was just text field. And I was like, this is what it takes to do a text field. Will 73 00:28:25,000 --> 00:28:55,000 It's ridiculous. And that was even back in the day when I don't even think I did my best work or we had the same standards that we do now where it would be even better, even stronger now for looking exactly like the way it was, um, material was scrappier back in the day too. I mean, like I said, it started with, I made some code. Can you share it and maintain it? Sure. We can do that for you to, um, my first project was actually building the IO app for some reason. We were like, okay, let's do that now. Will 74 00:28:55,000 --> 00:29:31,000 We're like, no, we do not have time to work on the IO app. . We're, we're, we power the whole world, you know, of front end in many ways. So it's, it's been quite a journey and I'm still very, very proud of some of the things that have come through dynamic color, user selected, contrast, the HCT color space. I think those are the greatest things that I've touched here because they are truly innovative under the hood. They use some of the most, um, you know, bleeding edge technology for dealing with the color. Will 75 00:29:31,000 --> 00:29:49,000 And they brought new concepts that hadn't existed before in order to personalize ui. And every bit of research that we have shows that people want personalized experiences, and that's what Google's known for, right? We personalize your search results, we're going to be personalizing your UI more. Yeah. And I think that's what my AI team is going to be working on. Liam 76 00:29:50,000 --> 00:30:13,000 Yeah. And I mean, people who have followed me on my Google journey know that I was saying years before I joined this company, that this was the place best positioned to realize a vision of true adoptive design. And I really see these making these subsystems relational in a way that abstracts them and allows them to respond to like unpredictable conditions is how we reach that future. Will 77 00:30:14,000 --> 00:30:43,000 And I also want you to ask yourself like, why haven't we reached the future already? Right? Like, it's Google, Google is gigantic. Mm-Hmm. , why have we still not gotten to a fully adaptive, um, software for all of the things? It's because it's still expensive. I mean, we can make helper code to help people get a tablet design, but then Gmail or whoever these giant applications have to sit and design it and make it Mm-hmm. and they have their own constraints and their own concerns. Will 78 00:30:43,000 --> 00:30:50,000 I think it's good for us showing people a way to move forward, but then they need to kinda like, take it on board and then apply it to their thing. Liam 79 00:30:51,000 --> 00:31:12,000 I'm thinking about the dimensionality of the interface that we're talking about two dimensions, something behind glass that you can't even physically touch and how you work, work more complexity into that system at a level of abstraction that's appropriate for millions or billions of people. Will 80 00:31:13,000 --> 00:31:37,000 Yeah. When I started on iOS, it was about how do we make this look like the real thing? , the skew morphs of it all was so much fun for engineers to work on. And I hope designers too, where you'd be like, this is a podcast app, so it's got to look like a podcast studio with the microphone and Yeah. And all these sorts of things. Um, I miss the fun of that, but I don't miss the look of that, if that makes sense. Liam 81 00:31:38,000 --> 00:32:17,000 But I have to say, it's interesting to me to think back on that now because I think metaphor has been at the heart of translating interface into a 2D screen since the beginning. And it's no coincidence that like the first creative conceit for a desktop interface was called a desktop . Hmm. Um, and I wonder, I wonder, I, I can't help but think that that's going to reenter the picture at some point. That, that the, the level of complexity and personalization that we're talking about, we'll have to take on a quality that helps people understand these concepts by referencing something that they already know. Liam 82 00:32:18,000 --> 00:32:18,000 Here's, Will Which Liam I guess which was the conceit of material . Will 83 00:32:20,000 --> 00:32:54,000 I, yes, it was. And that was really cool. Although, um, I think we actually did help ourselves by expanding that universe a little bit and stopping to be so literal about the material paper. Um, but I see in the future where this is going to come into play working on the AI design team, we're working with teams like Bespoke, which is this team that, um, has been featured on, um, in Gemini's, uh, kind of like press about how we can build designs at runtime that are mm-Hmm. completely bespoke to the user. Will 84 00:32:56,000 --> 00:33:32,000 Take this to an nth degree where you are seeing a UI that has been sending you experiments for years to figure out exactly what you like and exactly what you need. Because everybody also has very different usability issues. One of the things that we learned here is that some things like WCAG are actually constraining and not taking into account the fact that there's huge diversity in usability needs inside the world. But we could begin serving that to you. We could figure out through your usage exactly what you need and build you a UI that looks nothing like mine. Will 85 00:33:32,000 --> 00:34:04,000 Mm-Hmm. your experience with an application. The same application could end up becoming strange and weird and take into account the, um, indeterminacy of LLM work, you know, turning up the temperature if you want to see something that's avant garde almost, but works for you. And so we could end up with concepts that we don't even have yet that the computer tried out, that then we could codify into guidance that we could use in other places. Yeah. Because if something works for another person, it could be a good experiment for somebody else. Will 86 00:34:04,000 --> 00:34:38,000 This is why I came to Google . And then ask yourself, what is the value of the design system in that world as we march towards that, the design system is going to, sure. We're gonna start with just where are the components on the page? Then we're gonna start to, well, is it these components? Is design guidance about patterns, usability, creative direction, and expressing brands so that people still have a halo feeling of Google, um, made me feel happy. And so I care about it while I'm doing something that's pink and completely different from other people. Will 87 00:34:39,000 --> 00:35:14,000 And our design system's going to look the same way that they do in the future. Would they maybe for an end user product, like, um, an external company, but not for, um, Android or Chrome or something like, there's kind of like these huge possibilities that design systems blow up a little bit and change that. If we give control over to the computer to begin making design decisions, then we have to just think about how we're guiding the computer and not guiding other designers necessarily. Who knows? Will 88 00:35:14,000 --> 00:35:47,000 I mean, this is all theoretical right now. Like fun experiments are happening and it would have to be proven that it has value. But it's a great thought experiment to just think about, like, if I took away components, what is the design system? I think also about UI from science fiction, like from Star Trek Next Generation, and they had touch screens with all this stuff and all those beautiful sort of like, like jewel tone colors and how I think on purpose it looked nonsensical to us so that we wouldn't spend all of our time gazing into it, but instead staying with the people. Will 89 00:35:47,000 --> 00:36:20,000 Mm-Hmm. . But, uh, but also it seems like things that went further from the metaphor, it went further away from the, you know, the desktop like you were talking about. Maybe for some people things will iterate to that place. You know, we've done a lot of research here that showed that applications in China look completely different than applications here. The people wanted more density, more things, whereas us having to do with, um, you know, Latin and romantic languages need air around our letters and space in order for us to consume things correctly. Will 90 00:36:20,000 --> 00:36:46,000 So we have that, you know, Italian influenced minimalist design in a lot of Western UI that then doesn't speak to the rest of the world. If we kind of gave these experiments and let the computers do things, we might end up with new paradigms, new components, new sorts of things that just haven't been explored yet. And then we could end up with that Star Trek design because it organically happened and worked for somebody, which I think is really cool. Liam 91 00:36:47,000 --> 00:37:05,000 Yeah. And I think what's interesting to me about that, about seeing an interface that's so abstract that like someone is using it perfectly successfully, and I have no idea what's going on. Like it just looks like shapes on a screen to me. I'm, I'm, for some reason I'm picturing like a pink screen with like an orange star in a green circle or something, . Will 92 00:37:05,000 --> 00:37:09,000 That's 'cause you're looking at me right now. I think that's kind of, that's why I kinda look like Liam 93 00:37:10,000 --> 00:37:43,000 . Um, but I, I think what's interesting there is that there must be a recognition not only of what we're responding and adapting to, but the ways that those responsive patterns also affect the environment and the context in which they appear that there's a subjective impact on the person using it as well. That I think is gonna be really crucial as we carry out these experiments and see what direction we want to go. Will 94 00:37:44,000 --> 00:38:21,000 I miss putting in detail Easter egg love and attention into all ui the way that we did when iOS first came out. And that was kind of the standard that happened, um, as applications were becoming a thing and people found that the way to get noticed was by doing that sort of thing. And before like the markets just decided, oh, it's these five apps that you use on a daily basis. And when things were skew morphic, there was so much opportunity for that. It was like, oh, I could make something that when the, the billiard ball goes across the table, you can almost see the hair of the felt change. Will 95 00:38:21,000 --> 00:38:49,000 And we all loved that sort of thing where it's scrubbed a, a pull down, remember the pull to refreshes, uh, movement that had wacky sorts of things. It's a pizza app, so then it's a pizza being built, coming together and then being eaten while we're refreshing your UI sort of thing. I miss being able to put that sort of level of love into end applications. I don't work on end applications anyway, but I feel like there's, there's still great opportunity for us to bring that sort of detail and love to users and that they appreciate it. Liam 96 00:38:49,000 --> 00:38:55,000 Yes. You can tell when someone was happy making the thing that you're seeing. Will 97 00:38:56,000 --> 00:39:01,000 Well that's why you don't cry while you're cooking because then people will get sad when they eat it. Liam 98 00:39:02,000 --> 00:39:40,000 True . Um, I have to say, like going back to my point about, you know, my, my exuberant optimism about material and like where we're positioned, there was a part of me that kind of couldn't believe that we shipped the wavy progress indicator for like, playing music on Android. That that became like a system component. Like I think we are getting to a place where the joy of these components is, is allowed to be expressed. And to be able to do that at a place like Google with all of the complexity that you talked about is pretty impressive. Liam 99 00:39:40,000 --> 00:39:40,000 I think, Will 100 00:39:41,000 --> 00:40:27,000 You know, I'm gonna give some props to, um, my friends on the Android team, Lucas Din, uh, and his team worked on that slider because, um, Android definitely has a culture of wanting to go that extra mile sometimes. Um, Dan Sandler, one of the um, uh, directors of engineering over there, icon is, yeah, he's responsible for the Easter egg that gets put into every single Android release, um, which is so cool. And so Lucas was showing me the slider as he was working on it, um, to put it in there. And I don't believe anybody asked him to do that, but he saw and loved it and the creative direction was so strong that he said, this feels incomplete until I have it. Will 101 00:40:27,000 --> 00:40:55,000 And that's the sort of great partner that we love to work with over here is when that sort of thing matters to them that we're not like, trust us, this is great, but instead they're like, oh, I see that it's great and I can't wait to give this to people. Um, and there's also a tradition in material from several material releases of us showing people example applications of the creative direction that usually involve the music app. And then we have been called out on Twitter before Yes. For never building the music app. Will 102 00:40:55,000 --> 00:41:00,000 People not understanding, that's not our department. So I'm really glad that this was one of those times that could come full circle. Liam 103 00:41:01,000 --> 00:41:05,000 I have been among those calling us out before it was a Google , Will 104 00:41:07,000 --> 00:41:09,000 Where's my music app? Yeah. Where's my material to music Liam 105 00:41:09,000 --> 00:41:14,000 App? We saw the animations, but they're not on the phone. Where are they ? Will 106 00:41:14,000 --> 00:41:42,000 Um, I mean, people don't even really know all the things that actually don't even make it to ever being shown to the public, but they're sometimes incredible. Yeah. But just get cut for time, like in the movie or anything. I think I mentioned this to you in, in something I wrote to you recently. Um, material two was so revolutionarily gorgeous and they spent so much time coming up with, you know, a hundred ideas for it that I used to just sit inside the deck as a low little engineer enjoying myself, letting my eyes have a feast. Will 107 00:41:42,000 --> 00:41:52,000 And there was this one thing, um, on, uh, image treatments we called it, which was Oh yeah, applying filters. And how would you describe an image treatment? Liam 108 00:41:52,000 --> 00:42:10,000 Yeah, I guess my understanding is that some of them were kind of like shaders. You're like transforming the image visually in a way that could be like a duo tone treatment or like a half tone where it's composed of dots and the dots are changing size dynamically. Mostly like loading states or like that's Will 109 00:42:10,000 --> 00:42:10,000 The one Liam 110 00:42:11,000 --> 00:42:11,000 Ways Yes. Will Yeah. The dot loaded state one is the one we all remember because it was this idea of showing like a, like I'm uploading it in like a chat app because Google has a few of those. I'm uploading this image, it's going to take two seconds. So I have this treatment of OA of the image turned into, you know, Lichtenstein dots that are changing size to make a wave, to show that something is happening before it turns back to full color again to show me that it's finished. And it was pure inspiration. I mean, I asked the audience to begin a campaign saying, bring the image treatments. Will 111 00:42:44,000 --> 00:42:49,000 Google released the image treatments, the director's cut of material two deserves to be seen Liam 112 00:42:49,000 --> 00:43:01,000 . It does, it absolutely does. I will join that campaign. I'll sign the petition. That's, that's really where, where you could see the joy of the system coming out Will 113 00:43:02,000 --> 00:43:05,000 Absolutely made me feel better by whatever was going on. Liam 114 00:43:06,000 --> 00:43:16,000 Will thank you for joining me on Design Notes finally. This has been a fun conversation and I'm glad that we could finally have it on mic. I, Will 115 00:43:17,000 --> 00:43:28,000 It's been my pleasure, Liam. You know that you are absolutely one of my favorite people on the planet, so to get to Baskin your glory for a moment is really quite an honor. So thank you. Liam 116 00:43:28,000 --> 00:43:48,000 Thank you. You can subscribe to Design Notes on Spotify, apple Podcasts, or wherever you're listening right now. If you like this episode, leave us five stars and stay tuned for more interviews with the founders and stewards of Design at Google as we uncover new histories, perspectives and futures for the interface. As always, thanks for listening and sharing.