Adam Wiggins: Honestly, that was one of the core insights of Heroku when you talk about design insights was, developers or people too. Noel: Hello, and welcome to PodRocket, a web development podcast brought to you by LogRocket. LogRocket, combined session replay, error tracking, and product analytics to help software teams solve user reported issues, find issues faster, and improve conversion and adoption. Get a free trial at logrocket.com. My name is Noel, and with us today is Adam Wiggins. Adam is the co-founder of Muse, Ink & Switch, and Heroku. I'm super excited to talk to him. How's it going, Adam? Adam Wiggins: Hey, Noel, thanks for having me on. Noel: We've got a bunch to talk about. Want to talk about what you've been working on, your history, and one of your guys' recent podcast episodes you did on your podcast. But before we get into all that, can you just tell us about yourself, your role, your background, how you found yourself in this space? Adam Wiggins: Sure, yeah, I consider myself an entrepreneur first and foremost. But if I go to hard skills, maybe I'm a writer, I do product design in the sense of how it works. I do have in the very distant past, I'm now a gray beard by technology industry standards and started my life in the software engineering world. So I've done a little of all of it. But to me, the jack of all trades, bring it all together and maybe being very early stage building a business, I get to do a bit of all of it which I like. So ultimately, I like to build products that make people's lives better, or I like to build the company and the team that creates that product. Noel: The company you're at now, Ink & Switch, are you a co-founder or are you the sole founder? Adam Wiggins: I'm a big believer in multiplayer. Starting a business is such a hard path, for me at least. I got to do it together with people. All the really worthwhile things in my life have been collaboration. So yeah, my current venture is Muse, where I have four other great partners there and then Ink & Switch, I founded with actually some of the same folks that we did Heroku. And then Heroku had two co-founders for. So yeah, I always like to find that set of people that you, not only vibe with around a particular problem, but your skills compliment each other. And so yeah, I feel a lot of pride in everything I've created and hopefully, all of them reflect my identity. But ultimately it's never just about me. Noel: I found it interesting how from the space that they're in, Muse and Heroku feel like very opposite ends of the spectrum for me. But how did you find yourself going from Heroku, which in my head is technical hosting platform with quality of life tools that makes it easier for devs to deploy to more productivity software, if you'll forgive the connotation there. Adam Wiggins: Yeah, you could argue is a little bit of a career pivot there. Actually each of those three points there, Heroku, Ink & Switch, and Muse. So Heroku, yes, a cloud platform for deploying web apps, target audiences, developers. Ink & Switch is an industrial research lab with a focus on creative tools, and what's the next generation of productive computing look like. And there's a pretty broad set of research spaces that we explore there. And then Muse is an infinite canvas for thinking, and that was actually a spin out of the Ink & Switch research. We were looking into new document types. We think this canvas type that now with Figma and others is becoming much more common. We actually think this a really important next generation document type, and we saw an opportunity to build a specific application around that. And so for sure, okay, developer tools, research lab, productivity software seem like three fairly joint things perhaps. But to me, it's very logical. They all have this theme of tools. I like to help creative people build things. For whatever reason, I've just never been as interested in that end user thing of shopping, or social media, or something like that. I really am interested in, let me give you a tool that is, as you said, has great quality of life. Hopefully is joyful to use, feels good in the hand metaphorically speaking, but allows you to do your art, or your craft, or whatever it is, and create something. That's what gets me out of bed in the morning. Noel: I would speculate that people in this space, I feel like us on the LogRocket team, a lot of my coworkers and stuff are in that same head space. It's a privilege, because we're like creative people that like to create things and we are the target market for these things we're creating. It's like less, I shop on Shopify sites and that's fine. But it's like that doesn't bring me joy in the same way that I use some to them, this makes my life easier and it's really cool, I think, to be in that process. Yeah. Adam Wiggins: But I certainly know plenty of people in the tech industry that just find it completely boring to build productivity software in general. You think of, I don't know, Microsoft Office or something, and it's sounds pretty stodgy. Why would I want to do that when I can make games, or I could make that next generation social media product, or whatever. And that's great. It's good that people have different interests so we can all gravitate towards how we can best contribute to the world. Noel: Yeah, no, absolutely, absolutely. When you're vetting these things out, what are you thinking about in that target demographic? What makes a tool a joy to use and something that people will naturally gravitate towards? Adam Wiggins: Yeah, When you talk about tools, which again, you have this fairly indirect thing, you're helping someone create an end thing, it's as much an art as it is science. And actually I like that blending of those two things together, where there's this artistic just feelings based you. You know what you feel that what is it that makes this tool such a joy to use and allows me to create great things. But there's obviously a very pragmatic piece of it as well, which is a hammer needs to drive in the nail. Or the classic product trope I guess, which is people don't want to buy drills, they want to buy a hole in the wall. You got to solve that problem for them, give them the hole in the wall. But what in the meantime, can feel good when you're holding that drill in your hand? On our podcast, Metamuse, this is one of our top things, is product designed through this lens of creating tools. Whether it's for developers, whether it's for other kinds of creative professionals, word processors, spreadsheets, that sort of thing. What is that ineffable thing and how do you get to it? To me, it's an infinitely fascinating topic and it's part of why I've been able to devote a 20 plus year career to it so far, and I know we're near being bored yet. Noel: It's easy to curtail every innovation in the way that we've worked and just function into this umbrella of product design. Where do you pick out the slices that this is the area I want to focus on, this is where innovation can happen right now? How do you hone that down and figure out where to spend your energy? Adam Wiggins: Yeah, that's a great question. And especially in the research world, which is where Ink & Switch is, which is much more rather than thinking what's a startup that we can start charging money for in six months? It's thinking more where do we want computers to be 10 years from now, for example. And so you end up asking that exact question very broadly. I do start again, from the creative tools, and maybe the world that I want to see. So part of the impetus for us starting this research lab is that there's this shift in computing that obviously, everyone knows about. Which is basically mobile devices plus web/cloud that in the mid 2000s, up until mid 2010s, or now-ish, took computers from something that you really mostly used for... Call them productive tasks. People use them for games and stuff, but ultimately when you thought of a computer you thought, "Okay, I'm going to use this to do homework." Or, "I'm going to do this to do a spreadsheet," or something like that. And nowadays, using the broad umbrella for computing, which includes obviously, phones, and tablets, and things, and smart TVs, and so on. Mostly we use it for communication and messaging. We use it for entertainment, Netflix, and Spotify, and so forth, as well as just being connected to the world. That's what news and social media and all that stuff does. Basically every person in the world owns at least one computer now, and that wasn't true 20 years ago. But then that shifted exactly, yeah again, what the use cases are. And so now all the big platform manufacturers, Apple, Google, Microsoft, et cetera. Of course, the bulk of their revenue... Apple used to be a company that made the bicycle for the mine, tools for creators. That was their focus. That's where all their revenue came from. And then it happened 15 years ago that now most of their revenue comes from the iPhone, which ultimately is more of a end user consumption device/ Which again is totally fine, it's a great product. But then we stop and think, wait, who is thinking about making computers better for productivity, for creativity, for art, for science? Not really. It's a bit of an afterthought for the big platform makers and then it's hard to innovate. Individual apps can do things, but they're on these platforms that end up being more and more optimized for that. So if you start with that, let's say problem statement, and then you say, okay, well, let's imagine we had as many resources and as many smart people who have put their brains into giving us streaming video and touchscreens and things over the last decade or two. What if we had that same level of resource and thinking that went into productivity? And I like to even look at science fiction for inspiration. Take Tony Stark's lab from the Ironman movies, where he talks to his... Yeah, there's talking to your computer and stuff. That's a little goofy. But he's got multiple displays. They all work together. He can spin things around in 3D and that sort of thing. And that's obviously all Hollywood. But that's not what it feels like to do productive tasks on a computer. What could give us that feeling but be more real, for example? If you use that lens that opens just a vast number of avenues to pursue. Noel: Tony Stark doesn't have to worry about the file he saved on his Windows computer not being openable on his MacBook. That's not a thing. Adam Wiggins: Exactly. A real world Tony Stark lab, he'd be typing on the computer looking at the file, and he is like, "Okay, I need to take it over to my table here." And then he would spend five minutes fussing with the airdrop or something just trying to get the file open on the other device. Exactly. Yeah. Noel: Just a quick pause here to remind you that PodRocket is brought to you by LogRocket. LogRocket can help you understand exactly how users are experiencing your digital product with session replay, error tracking, product analytics, frustration indicators, performance monitoring, UX analytics, and more. Machine learning algorithms surface the most impactful issues affecting your users, so you can spend your time building a better product rather than hunting through tools. Solve user reported issues, find issues faster, and improve conversion and adoption with LogRocket. Adam Wiggins: Just to give you one concrete call design principle or something like that, I'm a huge fan of Unix, as many developer types are. And even though you don't think of the Unix as being a modern sense of a design user experience because it's primarily not graphical, it actually has these great design principles built into it that are designed to maximize the utility and the user experience of the people who are using it. And so for example, command line tools with their ability to pipe stuff in and out. And you get this composability where you can have lots of little small sharp tools, little tech space tools that you can chain together in an infinite variety of ways. That is a really cool principle. I think that principle of composability doesn't matter at all for end user stuff. You don't really want that from your, I don't know, calling a ride share. You don't want it to be composable, you just want to go through the steps and that's it. But for a tool you want that. And so coming at it from, okay, right now for example, we live in this world of really, of apps, and the apps are all pretty siloed. But the reality is creative people string together lots of different things. Whether you're talking about video production, developer tools, writing, whatever it is. You don't use just one tool, you put it all together. But it's actually really hard to get stuff between them. So one topic of research for the lab and an ongoing question for me is, how do we get that Unix composability, that was so awesome in the '70s and whatever, and bring it to modern tools? Noel: Yeah. Do you think that we have lost some of that composability in modern tools? A lot of these players are taking active steps to make it so you're in their ecosystem. A lot of people would point that finger at Apple. They intentionally make it so it's hard to collaborate across platforms because they want everybody bought in. Do you feel that that is a major challenge to overcome right now? Adam Wiggins: Hundred percent. And there's two forces that are causing that. One is a very crass business like Lock-in. There's value to Lock-in thing. Put that aside for the moment though and assume good intentions from the platform makers or whatever. I think the other reason is really for the target audience, if you're talking about an end user audience that are not power users, that are more in a consumption mindset, actually, it's desirable to have the computer be less powerful and less things that can go wrong. And mobile design is a perfect example. So here's a place where they got rid of things like minimizing windows or files. There's no difference between minimizing a program and quitting. There's not even a concept of quitting a program. That's actually great for most people who don't have any mental model for what does it mean to start a quit a program. They just want to bring the program up and then make the program go away. So in general, I think that's a big win. And I think taking files away, which the cloud and mobile mostly does, is a win for most people in most of these common circumstances. But again, you come to power users productivity, whether you're editing a video, editing, audio writing, programming, these kinds of tasks, you need that power of things like files, and windows, and you need to be able to connect things together. And that stuff, not only are we not investing in it really, we're actually neutering it because that serves the need of serving consumers. Maybe in my dream world there would be some new operating system, hardware platform that would rise up that would be purely for those productive uses and you no longer need to have that tension between your two target audiences. Now the reality is, it takes millions, maybe even billions to create a fully featured operating system computing ecosystem today. So, we'll see. But this is the sort of things that I spend my days thinking about for better, or for worse. Noel: I feel like when people are talking about product design, everyone just thinks about the UI plus some extra stuff, how one interacts with the UI. So I think it is interesting to zoom out a little bit and talk about these interfaces, how programs talk to each other, how users interact with multi-step processes to get stuff done. As Muse came to be, were you thinking about these kind of questions actively? Or was it more just, "Oh, this seems like a cool infinite canvas implementation. There's some really nice affordances we stumbled upon here?" Or was it more like high level thinking at the time? Adam Wiggins: Yeah, I'm a high concept guy. So everything I ever do is guided by those kinds of principles and thinking about, I guess, how I want see computers be, in a way that hopefully not just suits my interests, but also I think is potentially better for humanity in the long term. Again, for these kinds of pursuits like art and science, more than maybe some of the other things that computers could do for us. So yeah, when Muse came out of the research in the lab, so again, Muse infinite canvas for thinking. So it's basically on iPad or Mac, you can put lots of media on a open two-dimensional canvas, you have these nested boards. But I see this as a chance to take some of these ideas but not need to rewrite the operating system or change the hardware. Obviously, Apple has the best hardware in the world and while have some complaints with the operating system, at least within the window that we are given, we can try to make how we think productive computing could or should be. So for example, Muse has a concept of cards that you move around on a canvas essentially just objects. But very much, they're heavily inspired by files, which is you can put... Well, in many cases you put actual files like PDFs, or just an Excel file, or something like that. But also a link, or text, or images, or video, all of those things can go in there. But very much like the Unix, everything is a file thing. And Muse everything is a card. But now this is the modern version of this. You can move it with your finger on the touchscreen or with your mouse on the computer. And you can resize, and you can duplicate, and you can link them together, and you can have that every card behaves the same way. Even a video and a PDF are very different in some ways, but actually, you manipulate those objects the same way inside of Muse on this open canvas that also offers this infinite nesting that is sort of like a hierarchal file system, but is much more visual and spatial. So we're sort of saying, "Here's some things that are great about productive computing in the past. Let's use the power of what computers can do now, what they couldn't do even 10 years ago, in terms of rendering this canvas, or the zooming interface, or just the ability to render high-res video, or render render high-res PDFs and all that sort of thing." In a file browser, you typically have to maybe you get a blurry little thumbnail and there's a file name. What is this again? You double click it. "Oh yeah, okay. That one." You close it. In Muse, there's no concept of that. Everything just appears and is rendered. You never open and close. You just zoom freely and navigate around between them. So very much it is a chance to... We saw sort of a use case and a specific need that sort of existed in the world that we thought we could serve. But we're taking these big ideas about what we think computing can be, and trying to bake them into the box an app provides you. And there's a lot of limits to that and sometimes we're frustrated when we hit the edges of that. But we'll take it as far as we can within that, so that it's a thing that real people can use now as opposed to a research prototype... And not something that could be part of your life. We have to work within the status quo of what computing is today. Noel: Sure. Do you think that hierarchical structure, like infinite nesting of data, do you think that that's an inherently useful abstraction all the time? Or do you think that that is one that older generations think about from these days of navigating file systems, and stuff, and they just think, oh, data information on a computer is always in folders and I can infinitely nest them? Do you think that that is inherently a human way of organizing information? Or do you think that we've just stumbled into it? Adam Wiggins: Yeah, that's an excellent and nuanced point. So this is actually a topic that a lot of people in the human computer interaction, that's HCI field of academic research have looked into. And certainly there are different types of people. So developers, for example, tend to naturally think in hierarchies. It's not even completely clear which way the causality goes. It may be that the kind of person that naturally thinks in hierarchies is more easily drawn to programming or maybe it's just the act of programming that gets you thinking in that way objects that contain other objects. It is the case for sure that people get confused by hierarchy file systems in the classic sense, "Wait, where are my files?" There's some really interesting research here. There's a great book called Managing Our Digital Stuff, that looks back over a lot of experiments over the years, including BOS had this tagged file system. And a lot of people also have the point of view of, "Well, let's just make it search." Google made Yahoo hierarchal organization irrelevant. Why shouldn't our computers just be search? Things don't live any place. But if you look actually at the psychology of people, and this includes power users, who are not necessarily developers, there is something really powerful to having a file, or an object, or a document, whatever you want to think of that as. It has a place that it lives. It seems to come from our natural spatial thinking and the way our brains evolved. We evolved to navigate our environments. Our memories are often very connected to that. There's this whole world of memory palaces that is all about essentially visualizing physical places you've been and using that to remember things. And so, there seems to be something to the file that's on my desktop or the file that's in a folder on my computer. I understand that in a way that I don't understand something just being in a tagged file system floating in the ether somewhere. And I would argue actually the iPhone and, in general, mobile operating systems have also embraced that. It is a spatial setting where your stuff lives in a place. I know that icon is on the third. And yeah, there's search and stuff. But ultimately there is a sense of placeness of your home screen and your apps. And I think that actually is a really powerful foundational thing. And so, Muse is the power user version of that, which is you have this canvas that can stretch in both directions. You can totally freely place things. And then you have the boards are nested, you can think of it as nested whiteboards or something. That is a more sophisticated thing. It's a harder to grasp. It's a more challenging mental model compared to say an iPhone. But we're trying to grab the best parts of that spatial reasoning stuff. And again, assuming this is for powered users and people who want to invest in their tools, not necessarily totally casual user. But we're not necessarily expecting you to be a programmer or expecting you to be a gray beard like me that remembers the hierarchical file system. We're trying to get the best parts of those, but bring it into the modern and a more intuitive application. Noel: It's always struck me as a little odd. The desktop on a traditional PC, has this spacial element. You see people on their computer put folders in specific places on the desktop. And it's always been odd to me. Why is that the only place that level of abstraction exists? You get that one layer deep at the top, that's the only place you can do it. You can do it on Mac, you can move folders around in the windows as they open, and I think it saves it. But I don't know. I don't feel like most people do that. Adam Wiggins: Yeah. Yeah. Noel: Mobile does embrace it. People have folders. I mean, I'm a psychopath and I like search for app names every time I open them. But again, maybe I'm that weird programmer. Adam Wiggins: Yeah, I do use search myself quite a bit on my computer, on my phone, et cetera. Weirdly, I think the search actually works better if you know the app lives someplace. Noel: Yeah, absolutely. Adam Wiggins: Yeah, I don't know, maybe the web is different because it's all human knowledge. And so I think your own stuff. Maybe if you even think about objects you own that are in your house or something like that, just you keep track of mentally where they are. You don't need to do that with everything that's on the internet. And again, that's a difference between, there's my record collection or my Spotify playlist, and then there's every song the humanity has ever invented. And for some reason, for the first thing, it seems like some sense of it has a place that it lives spatially, even if it's just a position and a playlist is good. And then for all music that's ever been created, it doesn't make any sense, just search or whatever. I don't get any value from, or any mental affordance by having not live in a particular place. Noel: I think music is a good example. People's mental organizational model is going to vary wildly based on the person. So if it's not yet indexed into their model, it's going to like anyone trying to project like, oh, we'll do it by, I don't know, year, and then genre, and then beats per minute. It's not how some other person is going to think of music at all. Adam Wiggins: Well, and now you get into a whole other interesting area of HCI and sort of the tools for thought space that we're part of. Reaching back to Doug Engelbart who thought a lot about this in the '60s and '70s, which is the idea that if we have more and more of these digital spaces, we need to organize them the same way libraries organize. Should there be someone in an organization whose job is to figure out exactly how to catalog things, whether that's music, or documents, or your team wiki, or whatever. Honestly, I think that stuff is all really totally unsolved. We have some good tools. In many ways, our multiplayer technology is still pretty primitive, and our search is still surprisingly basic. When Google search is good for sort of all human knowledge. But when you come to let me do a search that's a subset to things that are part of my personal stuff or my organization, then it gets a lot worse. So I think these are still very active problems that we're working on. And by the way, this is from Muse, which started life as a personal tool, a space for personal thinking through these open canvases. Now we have a teams product. And that massively increases the utility because now it's shared whiteboard with your remote team, and you can work together, and develop a document, a design spec or a technical architecture, or a planning doc, develop that asynchronously over time, or together in real time. But now, you're in this spatial setting and everyone has their own opinion about how things should be organized. And some people like to be really tidy and things are organized, laid out like this. And other people don't want to be tidy and they just throw it wherever and "Hey, I'll search for what I need." And yet different ideas about exactly how it should be sorted. So we're only at the very beginning of figuring out how, especially when you think of pure remote teams, how it is that are vast space of information, and I'm not talking about the whole internet, again, I'm talking about your team space, or your personal space, or your work group space that can be organized, searched, accessed, kept fresh. To me, it's a really exciting space to be working in because there are relatively speaking, so few people, and teams, and designers who are thinking about this. But there's the fast open space that I think is really high impact for humanity. Noel: When you were thinking about multi users interacting on the same spatial canvas, did you guys ideate around should each user have their own visual representation, or should it always be one shared one that there's some reconciliation that happens to this is the master, but everyone else can have their own unique version of how the data's organized. Did you guys toss that around, play with those ideas at all? Adam Wiggins: Yeah, funny you should ask about that. That's a whole track of research at the lab, which is I think, changes over time, especially... Certainly for individuals, but in a group setting is a really rich area to develop. And this is actually a place where I think developer tools are way ahead of everybody else. The get GitHub, poll request, diffs and patches, comment threads, basically code reviews and discussions on a poll request where you're essentially commenting on this. You're commenting on the change rather than on a particular snapshot in time of the code. And you can reason in the form of this timeline of changes. This stuff is all, I think, really powerful. And one of the things we believe in the lab is that a version of that workflow should actually be part of every productivity tool. Writers sort of work this way with Google Docs suggest changes, lawyers sort of work this way with redlining. The tools are really clunky there by comparison to the get developer world of things. And you can make the same thing for video production, audio production. Almost anything where you have more than one person that needs to work on a work product that merging and changing over time and understanding what we're doing together. I think that's just a really important area for software. So then coming to Muse, yeah, so there's obviously just the basic multiplayer of merge changes together in a reliable way. Be able to see, have presence and have a real time thing if we're all in a document together in a meeting. But also asynchronously, I can come online and I see someone's done a bunch of changes, in that while I was away, and I can check that out. So even though Google Docs may have set a precedent from that now 15 years ago. Honestly it's still really hard to do that. There's no good off-the-shelf solutions. Most people roll their own. That's sort of almost like the table stakes, just make it possible to work on a shared document. And even that is very far from a cleanly solved problem for the industry. I think that is ideal in a lot of ways. We can each sort our own spaces the way that we want, and then put it through a lens that lets us see it in the way that we want. There is also a lot to be said about the power of we can see the exact same thing. This is why screen share is really useful. And even to this day, even with something like a Google Docs, Figma, whatever, multiplayer products, sometimes people still screen share. "Okay, I'm going to put up the meeting notes so you can see where I am in the document and we can all make sure we're seeing the same thing." There's quite a challenging trade-off there. But yeah, I think again, if we think way far out, which I always like to do, there is this path from right now we're trying to get just the, we're all looking at basically the same thing. Technology, we here, I mean as an industry, make that work pretty reasonably well. But I think once you get past that, you start to get into, okay, obviously there's basic things. "I don't know, my vision isn't as good. I want the font size to be bigger." Okay, well how does that change the flow of the document? How does that change the scroll position? Can we sort of see the same things even though my exact layout is a little different? And that's a very basic change. Obviously, you get much more. You can imagine someone's looking at the grid view and someone else is looking at the facial view, for example. Does it get confusing fast? But I think there's a positive version of that, which is in that utopian future, however far out it is, that we can create the views we all want on the same dataset. We may even be using different tools. We may be coming at it from... Certainly something we see on the news team a lot, where we're creating a space where many people from different disciplines can come together, a developer, a product designer, a marketing person. But they all have different preferences. The developers are like, "Can't we just write markdown and put this in a folder and Dropbox or whatever, or on GitHub?" And the designers are like, "No, no, no, let's just do it in Figma. It'll be fun." And people are like, "Oh God, Figma is so complicated." The marketing people are like, well, "Look, let's just discuss it all in an email thread." Each discipline has the tools that they love. But I think in the, again, that utopian future, it'd be something where we can have some shared data set. We can all use our own tools, our own preferences for how we view it. But be confident that we're not working across purposes because we're either using different tools that aren't compatible, or we're looking at the data in different ways. So we are literally not seeing the same thing, and therefore, we can't have productive collaboration. So it's going to be a long time before we reach that utopia. But I do believe it is possible. Noel: You noted there's a subset of user specific traits in a certain display. I think you mentioned font size or being able to view certain colors. So there there's a whole subset of things here that people may want to view data differently. Do you think this abstraction we were talking about before of hierarchies and data, do you think that that one is one that will always need to be inherently shared for this collaboration to be effective in this utopian world? Or do you think that even that, the way that the data is structured in a hierarchy, that can differ between users on a project that is being worked on? Adam Wiggins: Well, I do think that for sure you can have some people that care about or get utility from a hierarchal organization and others who just don't care. So for example, using Muse today, a lot of teams use it for planning sessions, for example. So there's some people on the team who are maybe the more power users. And they really care about, "Okay, we've got all our planning boards organized here. We've got a link out to our master plan for the year. Another link out to some ongoing projects. And I'm going to connect them all together in a way that makes it very possible to get a bird's eye view of everything we're doing." Then there's other people on the team who are just like, "I don't know, whatever. I just want to jump in the shared document for our call. And I don't care what it is, just give me a link." And they literally just aren't even aware of the hierarchy. They just copy a URL from the board, you paste it in Slack, or email, or whatever, they click on that link. And they're just there. They contribute to it, maybe drop a few comments, drop in a few files, whatever it is, and then the session's over. They just close the window. So I think there's certainly room for that, which is the degree to which you care about the information architecture can just vary a lot. Now having really different architectures per user, that's incredibly interesting. And actually it's something we've gotten into. We just did a podcast episode actually on linking, where we talked about the history of that. Obviously, the web, and wikis, and all that kind of stuff. As well as in this tool for thought world, you have more of this linking backlinking thing that you see in Notion, and Rome, and some of these types of products. And when we added linking to Muse, now that breaks the spatial model. And now you can, if you want, just have a completely different slice of the data by having different links across. Also gets really interesting if the permissions are different. "I have a link from something on one team I'm on, another link to something that's in my personal space. I actually want to put those side by side. But maybe someone who's in that same board with me, they're not really going to be able to navigate into this thing that sort of is in my personal space." Again, lots of potential for giving you all kinds of malleability of your own environment, and your own data, and be able to see the information, and the sphere of knowledge, and the way that works for you. But also lots of room to be incredibly confusing. I mean we already had that problem with Google Docs, which is just you share the link and it's like, "Wait, no, I'm not shared on that. Oh, did I not put it in the right thingy thing?" You're already getting confused in just that very basic thing. I do think it's solvable. And this is why it is such a rich, fertile territory, I think for people who are interested in this slice of product design, this how it works, and this creative tools thing, I will easily fill a career working on these problems. Noel: You've mentioned your podcast a couple of times. I recently listened to the remote work episode. And I think there's some interesting play here. You talked about how in dev tooling like Git, we have really good ways to discuss changes as they're happening. We don't have that elsewhere. And I feel like maybe that's why engineering has, I think, converted to these largely remote teams a lot more easily than other fields have. Do you think that that slow meticulous capacity to comment and reflect on every change, do you think that that is a unique privilege to engineering that other fields can't always afford that meticulous check every change, make sure everything makes sense? Or do you think it's some kind of middle ground there that can be walked? Adam Wiggins: I would frame it a little differently, which is it worth being that meticulous? So for example, I mentioned the other great success in the changes as a first class thing is the Google Doc suggests changes. So most professional writers I know just love this. If you're working with an editor, copy editor, you can suggest changes without changing the core thing. And then you can comment on those changes themselves, which is commenting on a poll request. It's much lighter weight. You can do it really quickly by just clicking your cursor and start typing. You can resolve it and basically accept the change by just clicking a checkbox. So compared to the whole developer flow, it's way lighter weight. But I think it's the right amount of complexity for the needs of that audience. And I would argue there's probably a similar thing for almost any other creative field that you can think of. And each tool maker in that field would need to find what that is. Now, I think there's two layers here, and this is what's making it an especially hard problem, and why I think of it as more of a research problem at this stage rather than something a commercial company or a startup could really go after. You have the infrastructure and get provided that in the developer world. Now, obviously, I'm a fan of... I used a version before that and go back in time. I've used quite a lot of revision control. But the decentralized revision control, this includes Darks, and Git, and Mercurial. The Git for whatever reason, just managed to be the success story there, created the infrastructure. Now the user experience and is certainly the concept of poll request and that sort of thing. That wasn't part of it. But it was the plumbing. How we can talk about ship these changes around in a way that's very efficient, and quick, and individual developers can reason about branches, use it in their own work even if you're not collaborating with others. And then you layer the whole GitHub world of things on top of that or GitLab as you prefer. So I think there's a similar thing that probably would need to happen to see this innovation with changes in other kinds of productivity software. Imagine trying to do this in video software. This is on my mind right now because a colleague of mine were trading some final cut files. And it's like even just making a video, saving it to a Dropbox file, and getting someone else to load it is incredibly difficult. They're just not made for that. They just assume exactly one person is going to work on it. And what we'd really like to do is, I'd like to be able to send a patch. I'd like to send her a patch which says, "I think we should add this in this here. And then we comment on those patches." She says, "No, that's not quite right because we solved this someplace else." You can't do that. All you can do is at best send a whole video file, and you quint at it, and scratch your head and go, "Okay, what changed here exactly?" And that's the state of the art in most productivity spaces. But I think it'd be pretty hard to just go indirectly put that into a video tool. Maybe you could do that. But you really need the infrastructure. How do you do diffs and patches, and change sets for video? Which is probably a technically challenging problem. And then what's the user experience on top of that? What do video producers and editors need? Do they need something that's pretty lightweight, more like Google Docs? Or is it a little heavier weight? Or how does the audio and the video changes in a leave? Can you do them independently? Is a whole host of user experience questions. I hope some team out there is already working on or someone will be inspired to. And again, for every domain, architecture, medicine, whatever it is, there's going to be a version of that, I think. Noel: Yeah. Do you think that this shift to remote that we've gone through past few years, is forced that a little bit? Or do you think it was primed to be happening regardless? Adam Wiggins: Yeah. Well, I guess, and indeed, coming back to your note about engineering teams could sort of go remote first. And there's probably also a proclivity there, which is, I don't want to stereotype too much, but if you think of the people that don't want to go to an office and interact with their colleagues, rather stay home in front of their computer, probably engineers are going to come to mind more than, I don't know, salespeople or something. There's obviously just such massive utility and remote work, and that's both for the person who, let's call it the worker. You have flexibility in your life, you regain that compute time. If you have kids like I do, then that it's a huge, huge thing to have that flexibility in your life. For the employer though, it's amazing to be able to hire from the global talent pool. And I mean the potential economic value that could be unlocked if we really embrace remote work. We basically haven't yet, from my point of view, we've sort of dipped toe. But a world where any company could hire anyone from anywhere in the world with ease, both from a legal perspective, a cultural perspective, and a tools perspective, that would be quite something I think in terms of... I think we would see that in our global productivity numbers. Just a hunch anyways. There's a chicken and egg thing there, which is when the tools are good, then it's easier to think, "Oh, I can make this work." And then especially if you're forced to, whether it's because of circumstances in your life, whether it's because of something going on in the world like a pandemic, and you find it works, and you go, "Oh, actually this works great actually. Why not do this all the time?" It's easy to take this perspective from a whole bunch of levels that developers are special, "We're smarter than everyone else, or we're nerdier than everyone else, or we're more introverted than everyone else." Pick something like that. And I think there's something to that. I mean, honestly, that was one of the core insights of Heroku when you talk about design insights was developers are people too. They are not as different as maybe developers themselves to think, or non-developers like to think as they look in and try to fathom what weird magic is going on as you write code into a glowing terminal. But I think that in many ways, there are these aesthetic differences. But I think in the end, most productive creative knowledge work is more similar, has more in common, and is just different in some details, is the way I think about it. So I would feel that developers tend to be more out in front on some of these trends for a whole bunch of reasons. Maybe this is even an opportunity for finding your next business opportunity, which is look at what developers are doing today and think about how that same thing could be brought to other spaces and productivity. Noel: Yeah, absolutely. I mean, I think, yeah, devs again, have the privilege of, "We're the ones working on the tools right now because we're the only ones that can," more or less. So it's like the tooling for us ends up being nice because we're in that space all the time. Adam Wiggins: Yep, yep. Noel: Is there anything that you're excited about more broadly, in the product design productivity space in the coming year? Or even just Muse in particular, anything on the roadmap you're excited about? Adam Wiggins: Yeah, well, I think that the canvas category that I put ourselves in, infinite canvas, I usually put it, I think it is having a little bit of a moment. Apple just came out with their infinite canvas thing called Freeform not too long ago. Lots of other great companies like Obsidian added that as a feature. More and more companies are making either as a feature or a full-fledged product, something that's this open canvas. And of course, you had that multi-player aspect. So it's weird to say that I'm excited about something in the near future that's actually the same thing I've been working on for the last several years. But in a way, it feels like it's close to a tipping point. Now, whether our big business and product will be able to capitalize on that or not is to be seen or that's up to our team to execute on. But again, a huge part of why I'm doing what I do is that I want to see the world and the computing world be different in a particular way, and I think that the canvases is under-invested in document type, is something that was why I was excited to start using the first place. And it feels like it's starting to have a moment. So that's probably top of mind for me right now. Noel: We'll get a bunch of links in the show notes, so listeners can go check it out. Thank you so much for coming on and chatting, Adam. It's been a pleasure. Adam Wiggins: Yeah, thanks for letting me ramble about philosophy. Noel: Oh, no, it was great. That was awesome. Thank you so much.