Future of Coding - Building Universe: Joe Cohen

Episode Date: December 13, 2017

Like many of us, Joe Cohen fell in love with HyperCard. Three years ago, he founded Universe to re-imagine HyperCard for the modern day. In this interview, Joe walks us through his initial vision for Universe, and the pivots along the way. It's a refreshing story about balancing pie-in-the-sky vision with shorter-term customer needs.  You can find the demo videos that Joe references here: http://futureofcoding.org/19-building-universe-joe-cohen.html Most importantly, you can download Universe for iPhone here: https://itunes.apple.com/us/app/universe-build-a-website/id1211437633Support us on Patreon: https://www.patreon.com/futureofcodingSee omnystudio.com/listener for privacy information.

Welcome to the Future of Coding. Today I have Joe Cohen on the podcast. So Joe is originally from Brooklyn, New York. He studied at Wharton, which actually makes us both University of Pennsylvania dropouts. Joe's first company was called Lore and it was in the education space, and he's currently building a company called Universe, which we're going to hear all about today. Joe, thanks so much for joining me. I'm really excited to be here. You've gone through a number of really interesting pivots with Universe, which I'm excited to

have us walk you through in a minute. But before we get to that, could you explain a bit about what you initially set out to accomplish with this company? Sure. I started Universe because I became increasingly convinced and passionate about making a tool that allows everybody to create the Internet. As the Internet creeps into our lives more and more, it sort of influences every function of society. The people who get to create it often are creating the world around us. And that world is only as interesting as the diversity of its shapers. And so five or so years ago, I started thinking about how we could design tools that allow people to create the web.

And that's taken on many different forms. It's a very ambiguous and sort of broad mandate. But at the high level, the mission is to empower all kinds of folks to create the world around us online. And so that's sort of the mission. As you may have guessed, you have the honor of being the first non-technical guest on the show. I'd be curious to hear about how that perspective affects running a company like Universe, which from my perspective is like a programming languages company.

Yeah. It's funny that you say that. It's funny that you consider us a programming language company. I don't think about it that way. I don't stay away from that kind of language because, A, my imposter syndrome kicks in. Like you said, I'm not a programmer. And B, I do think that it's intimidating to others and to our users. And so I think there are underlying factors that would warn us being

called a programming language, but I'll let you call it that. Yeah, I'm not technical. I'm not a programmer. I don't write code and I don't like writing code. With. Yeah, I'm not technical. I'm not a programmer. I don't write code, and I don't like writing code. With that said, I've been on the Internet since I'm 10 years old, and the Internet totally changed my life. I grew up in a sort of conservative Jewish community, and my world was really small. But getting online really opened me up to the world. And I soon started to create websites and just engage with other people and my interests online. And it completely broadened my perspective. And ever since I've started using the Internet i've i've sort of wanted to create

a tool like universe uh in that i want i wanted to make a tool that made creating the internet as easy as browsing or using the internet and i realized that as technology becomes more abstract and more useful to everybody, which has been the progression since the invention of the personal computer, what happens is that the domain of creating in that environment also becomes more democratic. So the media and the general narrative tends to concentrate on the sort of utilization of these tools. We use Excel, we use these programs, but not necessarily the creation of them. But the reality is that as we hit these sort of user interface frontiers, we also have the opportunity to make the creation tools themselves more human and more democratic. And I think that the phone was far and away the biggest leap in usability from a usage perspective

and thus presents an opportunity to do the same thing when we think of creating. But I think today it has still been rather limited. And so for us, we think about how we can make the phone something that everybody can use to create the internet. And we imagine a web that is more, like I said, represents all kinds of people, whether that's artists, teachers, small business owners, people who don't use a computer, but engage with the world, add something to the world. We think that an internet that represents those people, that is created by those people,

is a more interesting and a more powerful internet. And I think that if you look at tools like Twitter, we've seen the ramifications of what happens when you allow everybody to write 140 characters on the internet. I don't think we can even fathom what happens when you allow everybody to create the web itself. And so I think that if we can make that happen, we'll unlock a wave of creativity on a scale that we can't currently imagine. I could also talk a little bit more to return to the question about why this sort of personalizing and all of that. So in a sense because the question was how does my not being technical uh play into it i i think that um again like my experience here as a non-technical person guides guides the way i look at it on the

one hand i'm a technologist i i keep up with technology i design technology On the other hand, I try to retain the perspective of someone who to make with software in a way that wasn't ever possible before. So, for example, on Universe, we have people creating what we think of in the UI world as navigation, but they don't have parlance for it. They just sort of make it themselves. There's sort of this vernacular design happening. And that's a really cool thing. And why do they want to do that? Well, they want to do it because they've seen a website that does something similar. And so I really believe in this pattern of creation where, as humans, we see the world around us, and we want to make things that are similar to what we see and maybe push a little bit further. But we're anchored in the present. Because the web has proliferated so much, there are really good patterns that are sort of universally understood for what we can do with these technologies.

And because of that, there's a wide awareness of the ability for people to create this stuff and they want it. You see what I'm saying? Like if you see an online store, you know, if you sell stuff in the real world that you want an online store, you don't need to invent an online store because the notion of an online store is in the periphery.

And so you just have to say, I want that, which is a much easier process than a zero-to-one authorship. I definitely know what you mean. I see the, I guess, like the inverse of this when I teach children to code and they have no idea what's even possible for someone like them to make with a computer and they just look at me blankly. And then after they make their first few games, they're like, oh, now I get it. Once they get a sense of what's possible, they can go from there. So I guess what you're saying is that the internet has pervaded our lives

so thoroughly that every person has a sense of what's possible to make on the Internet. Yeah, because like the way I look at it is like, you know, if you think of the websites and the apps that we use as the architecture of the Internet, this world is now sort of developed enough that we have patterns. We have our ideas of what a building is, what a storefront is, what a street is. And we know enough as regular Internet users to say, oh, I want to put up a store here. I want to put up a building there. I want to put up a wall. I want to paint it red. Like we have these concepts, just like it would be fairly easy for a layperson to have an idea for a store and think and dream up a store. I think we sort of have that now on the Internet.

So I guess what you're saying is that as someone who, as a non-technical person, for you, it's very clear what the sorts of things that non-technical people would want to do with computers. Because it's the fabric, because we all interact with them all the time. Is that what you're getting at? Yeah, and I'll say, like, earlier in my career, I really felt the pressure to learn to code and sort of become a programmer. And, you know, like, I can write code. I'm very, you know, I'm not, I'm not good at it. Um, and, and you would not want me writing your, your code, but I felt this sort of pressure, um, over the years. And at some point I realized that I actually don't enjoy thinking, um, in code. I don't enjoy writing code. I don't enjoy it. It doesn't make

me happy. And I only imagine that that's actually true for perhaps most people, and that it's a very specific modality of expression. And so I actually changed my attitude on it, which was to really lean into the fact that this is not how my brain works. This is not what environment I thrive in. And to say, oh, can we create an environment that people like me can really excel in and sort of really push? And could we build a medium that was perhaps at some point as expressive as code, but in a way that was more accessible. Totally.

You definitely have my agreement that the coding we have now is not humane. And for most people, some people would even argue for all people, it's definitely not the most fluent medium for us to think in. Given that you're kind of like leaning into your dislike of it, I'd be curious to know if you've kind of gone deeper into like what particular

things about it don't jive with you? Well, I mean, again, so here my knowledge is limited. But what I'd say is I think text is a medium I love reading I love writing I love I love text but as a creative medium as a general reading a medium for me it doesn't doesn't stimulate me I find it very hard to focus on a task when it is a linear document and keeping track of what's happening in my mind that a visual representation is very difficult to do. That's a great criticism. Yeah. I think spatially,

I think about things as they are in the real world. So that's my main problem. It's not a criticism. That's actually the thing I want to say. I actually think coding as a medium works for a lot of people, but I don't mean to say it as a blanket criticism. It's just, for me, not the right tool. I mostly agree with what you're saying.

I think the nuance that I'm trying to point at is, is a point that Brad Victor makes in a lot of his essays, but in particular kill math, which I was reading recently. He says that the people who are good at symbolic reasoning or, and that applies to math or to programming are the people who are able to take the symbols and interpret them into this high-resolution picture in their head. So the difference isn't that you think in pictures

and programmers don't. It's that programmers can turn the symbols into pictures. Just speaking for myself, I'm able to turn the code into a higher-level picture in my head, and it took me a lot of time to get to a place where I could do that. So what I would say is if we were able to make tools

that visualized it in a picture, and I'm using the word picture, I think broadly, it doesn't have to actually be a picture, like something that's more visual, then I would have to do less work. Well, first of all, I wouldn't have to learn how to do it in the first place. And then even as a programmer or day-to-day, I would have to do less work. Well, first of all, I wouldn't have to learn how to do it in the first place. And then even as a programmer day-to-day, I would have to do less work. And then someone

like you wouldn't, you know, you could just immediately go to the visualization. So that's kind of, and to give you a specific example, there's one of my favorite tools in that particular space is this programming language, this JavaScript framework that's kind of like React.js. It's called Cycle.js. And they actually parse code, just regular, like JavaScript code written in this framework into a diagram with arrows. And then when you click on different buttons in your app, you can see the data flowing

through the diagram so with one glance you can immediately see how how a user interface is built up and and i think that's that's pretty powerful but it doesn't go the other direction you can't like move the nodes in the graph and have it have it come back together so let me add a couple things there i i hear you on the point that uh programmers should train themselves to visualize what their code is doing. But that also comes, and I don't think that is universally a bad thing, though. I actually think with that skill comes a sort of efficiency and speed, right? So if you do have the ability to do that, then the interface of code is actually, it's like a race car.

And I think that's sort of in the Engelbart spirit of building professional tools. I think there's a market for those kinds of tools. I mean, I know limited interacting with Git through the command line, it's more fun than using the GitHub client, the UI for it. You feel like you're, you feel much faster. And so I see the benefit there. I just don't think that's actually something for most people. The other thing I'd say is that not only is there this sort of visualization part, but

for someone like me, for example, I don't, you know, like if we look at a programming language, there are sort of abstractions of how the machine works. You know, when you start something like assembly and then abstract out and, you know, to they're still in the mold of how the machine works. And so you're sort of starting from the machine and building layers of abstraction above that. And we are not looking at it that way. We are starting from the human, and we're bringing the human closer to what's possible with the machine. And so I don't anticipate that we'll have an is to teach you how the computer works in a sense. And we're not actually interested in that. We want to allow the tool to be able to do things for you. figure out a more human way of explaining what's happening there even if we have to quote unquote rebrand it and and perhaps even limit its function um that's the route we're going to take yeah i

think that's um exactly the right idea divorcing the um interface to computers from the the way the way the computer actually works. That's really the main problem with programming tools or the main problem with programming languages is that they haven't been liberated from the minutiae of how the hardware actually works. And we're so far away from that. And I think that's why Universe is such a beautiful program because you're coming at it

from a totally different perspective. And that's, I guess, kind of what I was getting at with the question. Like you're coming from so outside the perspective of an engineer that your thinking isn't really bogged down by the metaphors that my brain is trapped in even when I try and get out.

Look, I think we're also very good as people we're really good optimizers. So if we start at a place, we'll find the local maximum. And I think if you start from a traditional programming language, even if you get to the local maximum of usability, I think that you're leaving a lot on the table and that there's actually a global maximum of usability that that that sort of in a that starts in a point that's sort of disjoint from the the lineage of uh

of traditional programming yeah well i like that visual because uh in in in like practice in order to find that new local maxima, you might have to start at a place which is lower than the local maxima that programming is already at. And then like slowly over time, climb back up a different hill, which is the peak of which is higher than the local maxima peak. Exactly. Yeah. Okay. So anyways, a lot of abstract conversation. I love it. I want to get a little more concrete so people can get a sense of what we're talking about.

So at the last future programming meeting, you described different phases of your company. So could you walk us through the early days and through your various pivots? Yeah. So I actually, I didn't even tell you this part last time, but this dream really started at my first company. And my first company, we made a product that was sort of like a website for your class if you were in college. So it was a place that you'd go to check your grades, download the files for homework, see the schedule, interact with your colleagues, etc. At some point, I became really interested in what the future of educational media looked like. And I specifically became interested in what the future of educational media on a phone looked like because it was clear that the phone was going to become the primary computer in our lives.

And as I went down that path, I realized that some of the ideas I was having were much larger than just for educational use, but rather that we could build sort of general purpose creation tools. I became really interested in that. It didn't end up making sense to pursue that within my first company, Lore, but after we sold it, I became obsessed with this mission. And I discovered Apple's HyperCard, which I think is probably familiar to this audience, but a relatively obscure product uh for most people

so quite a question for you just to jump in there did you ever actually play with hypercard because i've never had everyone talks about it but i've never played with it yeah yeah so i was able to get at the time uh i was able to get an emulator so i could run it on my mac and i used it extensively can you tell us a bit yeah tell me a bit bit about, I've only heard about it in abstractions. Can you give us some specific examples about what made you fall in love? Yes. Okay, so you run Mac OS 7 in an emulator, and it's this tiny screen running on my big, huge retina screen.

And basically the way it works is that you create a new stack of cards, and a card is a landscape-oriented blank canvas that you could then put elements on. And an element could be text. It could be a graphical element like a shape. It's all black and white, by the way. And you could drag these elements around on the screen. And so you could think about it as starting as like a PowerPoint-like interface or keynote-like interface. So you have blank canvas, adding elements to it. Now, let's keep in mind that this launched in, I believe, 1987.

And it was intended to be sort of software sister to the Mac hardware. And it was designed by one of the lead engineers and designers on the Mac OS, Joe Atkinson. And so you place these elements on the screen, sort of just like a graphical authoring environment. But what was really cool is that you could then double click on any element and you could give that element behavior. And you give behavior by giving it, sort of attaching some code to it and the code uh was

written in a language called hyper talk which they wrote for hypercard and was very english-like um and so someone like me could totally understand it and write it um and and you can do all kinds of cool stuff with it like you could i remember one of the functions you could have it like dial a phone number so you could like make a keypad for a phone. Which I thought was really cool. But the other thing about HyperCard was that you weren't just making a single canvas. You were making a stack of these cards. And so you could link one card to another card.

So if you created a button on one card, you could link it to another card. And so it actually introduced the idea of hyperlinking, which then went on to influence Tim Berners-Lee, who invented the web. Oh, interesting. So I didn't realize that that's where linking came from. Yeah. I'm sure there were other experiments and similar ideas at the time. But when you look at it retroactively, the sequence seems to be that Atkinson invented linking, but it was all local. So you can only link to a card within your stack.

And then Tim Berners-Lee took that idea and married it with the internet and that you could link to any document on the internet which totally changed the use case right like the network itself you needed hyperlinking and the network for it to be interesting um but anyway hypercard uh it was a medium that was broad enough that you could make anything from a piece of art to a calendar application. And you could use the stack, like the index of cards, almost as a database. So it was just this remarkably versatile tool. And one specific detail I really loved was that in the settings for the app,

you could actually choose the level, the sort of the level of difficulty. There are five levels. And the first level is view only. So no editing. And you say, why would you want that? Well, because a lot of people made their apps and their content in hypercard so in the 80s if you like bought a a video game it might

come in hypercard format you'd buy it on a floppy disk and then you put it in your mac and it may it may run on hypercard um and so that was the base level Then you had, I don't know, I don't remember the exact sequence, but you could first create graphical elements, then you could create links and buttons, then you could add the fifth level of scripting. But there were a couple things I loved about that. First is that as a designer of software, that's how I think about great software. It has a gradient. So it doesn't have a low ceiling. It has sort of a low step and a high ceiling, right? Like it's really easy to get started, but it also allows you to go wherever you want to go. I think tools either have really open-ended possibilities, but they're really intimidating or they're really simple and just simplistic.

They don't allow you to do anything more than that. And so what I liked about it at first was that it had articulated its design through this. Like it was clearly designed for that leveling up. I realized, though, that like that making that a user toggle um sort of defeats the purpose i think they probably did it for like educational environments because natcha really popular in schools but you know it would be much more interesting if the software itself uh allowed you to level up when when you were ready or when you wanted to instead of like having to go configure it as a parental control kind of thing.

In any case, when I saw HyperCard for the first time, it felt like, aha, this is it. This is an environment that I could use to make stuff online. And that, well, HyperCard didn't really work because it was pre-web. It was pre-network. So if you made something and you wanted to share it, then we could sort of unleash a bonanza of creativity. That HyperCard was 30 years ahead of its time.

And that if we brought it back for the first computer that was really personal, then it would open up these frontiers. So anyway, I became obsessed with this product. And actually, throughout this exploration, I moved to the Bay Area. I actually reached out to Bill Atkinson, who's nice enough to invite me to his home, got to go to his house for lunch. And he told me about his dream at the time, what led to Hypercard, which was actually an LSD trip. And yeah, I guess, you know, sometimes good things come out of that. But, you know, he walked me through the vision and I just left super inspired. And at the time, you know, I'm an entrepreneur.

I love business. I had a bunch of ideas that I was thinking about. This one felt the most nebulous, the most hard to sort of fathom and sort of make concrete, but it also felt like I couldn't not do it. And so I decided to pursue it. And I called one of our previous investors, a guy named David Tisch from my first company. And David had actually programmed with HyperCcard is how he learned his program and he got it immediately even though like most people had no idea what's going on and david wrote check without us even having a product or anything and

that led me to start experimenting um and so i'll give you like an abbreviated sort of overview of the history of the company. I started the company over about three and a half years ago. For the first two and a half years, we were working on a different product from what we're working on now. The first product was all within the same umbrella of empowering everybody to create the Internet. You can think about the first product as us wanting to create a new web entirely. What I mean by a new web is a new artifact.

Instead of a web page, you'd have what we called a verse. And you'd have a new way of creating these verses, a new editor for doing that. So you could replace a text editor like you do for a web page with this new verse editor. And these verses would be shared not on the web, but they'd be shared in this new network called Universe. And so we're sort of building this little parallel network. And the reason why we wanted to do that was because we felt like the web wasn't ideal in many ways. And apps in their state at the time, remember this is 2014, felt also not like the right democratic medium. And so when I started the company, we kicked off a design exploration for what this new web could look like. And that actually led to three unique user interfaces, specifically around the creation tool itself. The first user interface was very similar to

HyperCard. It was a freeform canvas. And you could put elements on that canvas, and you could even program those elements. So, for example, you could put a square on the screen, and you can say, when this square is tapped, then move across the screen. And we built some really cool UI, and I can share a video of what this looked like. We built some breakthrough UI for it. At the end of the day, after about nine months of work, so a long time, it fucked. It didn't work. We started putting it in front of people, and they really just didn't get it.

And we realized that we'd been so deep in our hole, like so deep in it, that what felt like second nature to us was actually a complete fabrication of reality like we had built our own little world and it made sense to us because we made it but we lost sight of actually what made sense to someone who doesn't have that context um and so we actually abandoned that uh we took the learnings and we went the opposite way we said We said instead of it being a free-form canvas, we're gonna double down on this whole idea of cards and we're gonna make these cards very structured. So you'll have three types of cards you can pick from. You have an image card, a text card,

and you have a card with an image and text, that's it. But what was interesting about these cards is that you could link them to each other. And so this one we started testing much sooner than, it was, I think, like a month into it we started testing. And that one started getting, like, working, but it wasn't interesting. Ultimately, like, it was sort of like text pages

with single graphics and just we you know we started the company because we wanted this open-ended environment and this was not that and so I got I started designing I started it actually started as initially a new type of card um but I wanted to create like a card type that was uh easy enough as easy as the sort of previous cards, but was fundamentally open-ended. And started thinking about the problem in a bit of an abstract way. And I said, like, one of the reasons why dealing with a blank canvas is too intimidating is because your possibility space is infinite. You know, technically the pixels on your phone are arranged in a grid, but the pixels are so small now that you can't discern them. And so the effect is a lot of paper

and a blank page is about the most intimidating thing you can do, give to someone, especially a As I said, how do we make this possibility space narrower? How do we effectively add some constraints? How do we reduce the decision space? And I sort of had this idea to take the pixel grid of the iPhone and make it jumbo-sized, magnify it, like enormously. I said, imagine if we divided the phone screen into a grid that was almost childlike. So we divide it into three columns and five rows,

which would give you 15 squares to fill the iphone's 16 by 9 ratio screen and i tried a bunch of different um like grid arrangements you know four by seven etc smaller finer ones but three by five felt like the perfect medium it was just accessible enough um, but still, you know, enough control. And I said, what if in that grid, you could draw on it, you could sort of take your finger and draw over the unit, the cells in the grid to create a block. And so the idea started unfolding, and I started working sketch, and it started to get really interesting.

The way it would work is you drag your finger over the cells in the grid, and then we'd bring up this block tray. You choose what you want. It would fly into that area that you selected, and then you can choose a picture or write some text. And it became very clear to me that this was sort of the interface we'd be searching for because it was really easy to use and really fun, actually, but it was totally open-ended. It was a modular system. And I'm

sort of a student of modular systems or maybe a sucker for them and I wanted to create something that was like this that was sort of Lego like in its simplicity and its openness and it felt just perfect and so I just got really, really excited and we started prototyping and it felt like the interface that I'd been searching for and starting to think of these things. uh went back to the drawing board and decided that we were going to abandon the previous version of the product which is a defined type card one and instead we were going to um we were going to go with the grid and uh the grid was one part of it the thing that you were creating was called a verse so you can think of these verses like little mini canvases that are interactive. So, for example, you could make a little informational page. You could make a little game, actually.

And the elements that you put on the grid didn't just have visual media, but you could actually add what we call a script to it. And the script followed a format that was like Mad Libs style. So you had a sentence that said, when this block is tapped, which looked like a pill, it looked like a button, then do blink. And then we had a few options that you could choose, like, for example example move three grid

units up or take me to another verse and so that quickly became the most popular one linking to other pages but we started to introduce this idea of like making the screens not just easy to design, but also easy to program. Now, so that's just, so we've got the grid editor, the verse, and then we had this thing called universe, which was the place where you'd share these verses. And the first version of that looked similar to Instagram, so you'd have people in a network, and you could follow them, and you can get access to their verses.

And so that was the version that we shipped in August 2016, so a little bit over a year ago. And everyone who saw it thought it was extremely cool. We got featured by Apple. But at the end of the day, most people didn't know what to do with it. They didn't know what it was for. They thought it was nifty, but they never came back. And more importantly, they didn't really know what to do with it. And so we started iterating on that, and we started finding our niche.

And our niche ended up being sort of art students, actually. So people who are mobile tech savvy, not necessarily technical, but curious about exploring new mediums um which was awesome like we started seeing like an outpouring of creativity but it was very small and it started to become in a different way an exclusive community of creatives um i also at the time started to think about what i really wanted to be building and did I really want to be creating a social network? Did I want to be creating a place? Did I want to be designing something where our success is dependent on how much time people spent in our little world? And I started to doubt that. I felt like our incentives were not necessarily aligned with the well-being of our users. And that didn't feel great. Last year, after the new year, so coming into 2017, we realized that what we had wasn't working.

We also had some team changes. We sort of went from seven people to two people. But for a lot of reasons, we realized that the path that we were on was not going to continue to work and that at the core of it was our product wasn't resonating and that we were actually failing at our core mission. The technology was cool. The user interface design was cool, but at a positioning level, it wasn't resonating. It didn't make sense. And so we were failing to communicate our value um so we start to think about we start to question the idea of versus existing exclusively in universe because that seems to be the real

problem it's like these things are limited in value because to see them you need to be in in this new world like this new app and you have to go download this app to do it. And so we started to experiment with, what if that wasn't the case? What if verses were visible on the web? And as we went down that route, we started to ask, what if verses weren't just visible on the web, but what if verses were websites?

What if, instead of universe being a way to create a verse, it was the easiest way to make a website, the easiest way to make a website from your phone in under a minute with a.com, a domain, hosting, all of that stuff abstracted away. Not just the easiest way to make a website, but an environment where you could get started really simply, but the sky is actually the limit in terms of what you want to do with it because it's built on this open-ended grid. That started to become really interesting. We started prototyping with some of the technology for registering domains and realized that

we could actually do a domain registration instantaneously from the app and that we could actually use Apple's in-app subscriptions to process them. Yeah, so I think that's something I want to just pause there for a second because that's one of my favorite features of Universe. That's just like nobody's done that before. It's 2017 and you're like the only person who's figured out how to make it so you can purchase a domain name with your fingerprint like it's crazy yeah there's a bunch of work there and for us you know we charge for these domains to cover our costs that is to us that is a service

we're doing it because having your own customer domain is so cool. And so we want that to be as easy as possible to do. And so that's not a money maker for us, but it is a service. It makes the whole experience complete. And I think we talked about this earlier, um you know with the first version of universe that we ended up shipping two months later in march it was very limited we we we really ship a very bare bones v1 and this is the new version that makes websites the new web yeah so uh we we ended up we ended up realizing that we were on onto something with this new website builder approach, but

we wanted to get it out as soon as possible because we wanted to really iterate it and learn in public about what people wanted and needed. So the first version that we shipped was really simplistic. It was a fixed grid. You could only make one page site. There were very few blocks that you could add to your site. It was just very limited, and there were some big bugs. But the thing that we wanted to make sure that we had was that even if it was small, what you could do, not having a website to having a custom website of your own making

an own design on the internet in less than one minute with a dot com. That was really important to us. So we would cover everything from design to hosting to domain registration, et cetera, in an under a minute experience. So we prioritize that as opposed to like, for example, really going deep on the grid at the expense of, for example, handling domain. So we launched that in March and then it immediately started working. And since then, we've been growing pretty strongly.

And it's amazing to see what people are making with it. Lots and lots of websites have been created. We've been growing at a very healthy cliff. And we've been iterating the product rapidly. So we are over with the days of monolithic app releases. We release on a weekly, bi-weekly basis. We're constantly adding new features, constantly rethinking our user interface.

We are having conversations with our customers on a daily basis. In fact, up until just recently, I did every customer support chat with our users. And we're using that as like input for our design process, but it's addictive once you go down this iterative route. What's really cool is that when we first started working on this, we weren't sure if there was going to be a sort of limited appetite for what you people want to do creatively on their phone like we'd hope that there would be an ambition to create things that were really

interesting and big but we weren't sure that that demand would be there but what we've learned is that the overwhelming feedback for us is people want more. People want to do more with the universe. People want to be able to make more things. And that is super exciting. So we have a product queue that's just immense. And it's something that we could continue to work on for a long time, which just excites us to our core um and really think about how we can humanize the design creation development experience at every level uh so we're still very very early our product is still very small um but we are uh rapidly rapidly inventing and iterating yeah so it sounds like you're a real convert from the um you know waterfall development style to the more iterative development style and um and and that's that's a it's a very i think most listeners are familiar with both styles and in modern times you know most people are talking about the more iterative style. I think like a good, an example of a

company that doesn't do that, I guess would be Apple that, you know, they release a phone once every year. It's kind of, as opposed to releasing new phones constantly, like Tesla, for example, every new car that comes off the lot has maybe better technology than the car, a few, you know, a few, than a car that came off a lot a few weeks ago. And so part of me, like, clearly, I see the benefit, the benefits of continuous, rapid iteration, like you're doing now. But as someone who converted recently, clearly, you thought there were benefits to the, to the, you know, spending time in your own universe for a little while, trying to figure out what people need, almost like the, like the, henry ford uh you know if i asked people what they want they

would have said faster horses um like like basically if you're too iterative and you listen to customers too much you could be trapped in this local minima so i'm curious how you reconcile that um you know yeah having a big vision with um with also you know iterating towards something customers want so a few things the first is that i think that projects have different phases and that you can think about like the first phase of universe as being like a research phase and that is broad it's open-ended, it's exploratory, and there's a lot of virtue in that. It allows you to be very creative and very inventive. I think, so you're 100% right, I am a convert to this new way of building. And the reason why I am is because I think it

results in better work. And the reason why it results in better work is because you put something out and you learn, and the feedback loop is just much tighter. Now, here's the thing, though. We listen to our users. We don't do what they say necessarily. So we interpret what they're saying. Our job as designers is to use that as input to figure out the best solution and to understand what the real problem is that people have. So it would be a mistake for a company that's trying to do something interesting to just do what people tell them to do. If they're any good at what they do, they should be able to take that and understand

the core of what those people want and then come up with the best solution for that thing. And so that's one thing I would really stress. But I also want to bring up two points or another point, which is that this is actually only newly possible with iOS apps. So we're an iOS app. It used to be that submitting a new app on the App Store took two weeks for an approval. Now it takes a day. And with iOS 11, when you submit a build, it doesn't wipe away your ratings and reviews.

So there's no cost at all to adding an update to your release. And now at this point, most users have auto update turned on. So they'll get an update immediately as soon as it goes out. So those are enabling factors that make it possible to iteratively develop apps in a way that really you couldn't do for the first six or so seven or so years of the ios platform but that have been the case on the web for a long time um so ios development has become more like web development it's still a ways away the thing i'd say in terms of um like a company like apple uh versus a company like tes Tesla, I think that the orientation is fundamentally a hardware versus software orientation.

Hardware doesn't work in this fashion. It doesn't work iteratively. You have to get it right. You have one shot, and it needs to be perfect. You can't change it once you've shifted it. Software, on the other hand, you can change. And I tend to think that the world is going to look more and more like software in the future. And that if you have a choice between an iterative approach and a non-iterative approach, and the underlying technology affords iteration, that's always going to win. It's just a matter of whether the medium itself allows

for an iterative process. And hardware to date just doesn't. So with Tesla, what's interesting is that they're shipping cars, which they can't change post-faco, but they can change the software that the cars run. Now, I think you're going to start to see hardware companies that incorporate the iterative mindset because they're started by people by founders who have a software experience because once you develop software iteratively it's hard to do anything differently like i think movies are going to be made iteratively i think that you know what we typically think of as artwork is going to be made differently music like and we're starting to see this actually happen like look at kanye west's last album life of pablo like he started changing

that after he released it that's going to continue to happen in everything like whether that's a building in the physical world or anything else and so my my my sort of thought is that if it can be software it will be software and if it can be software it will be iterative. I definitely resonate with that. I, like in my own personal work, for example, I have so much trouble declaring like anything that I do done because I know that I just want to make, like, for example, I have a programming language platform that I put out and there's definitely, my dad asks me, okay, so when is it done? Like, well, what do you mean when is it done?

Like, I'm just going to keep making it better and better. And yet, like when I write like an essay, for example, you know, I'm supposed to make sure that like that's done. Why don't I just keep making that better and better? I'm totally with you that this mentality of continuous improvement for forever will permeate all the things that we do over time. Yeah. Now, I think that it requires us to be more disciplined as creators because sometimes a thing requires more in-depth invention and you really do have to go out into left field. I just think we have to know when to use which tool. But I think generally the iterative approach, because it's new relatively, is massively underutilized compared to the other

approach. Most things are built in these sort of monolithic ways. And so I think, heuristically, we're going to move much more towards the iterative approach. But once we have that capability, there are projects that require deep thought and time. And I'd also say that there are some projects that also deserve an ending. Not everything should live on forever. And that's just fine. But I think that's going to become our choice instead of determined by physics, for example. You see what I'm saying? Of course. Yeah. So like back in the days, you would have to finish a book because, you know, they have to like mass produce something because like they're physical books or like a video game. You have to like print it onto discs. You need to stop working on it.

Yeah. So I really loved you walking us through the different pivots of the company, partially because a lot of the moves you made feel like moves that I've made in the past and I'm sure I'm going to make in the future. In particular, what really resonated with me is how obvious your current iteration is, how obvious it feels in now in retrospect, like, you know, it's a website builder for your phone. I think like, like Squarespace on your phone, or like a mobile for Squarespace is like a very quick way to describe it. So given that it feels so obvious now, do you ever look back on it and think, like, how did I miss this? Or because I personally do this all the time i'm

constantly coming up with messaging for my products it makes sense in my head that doesn't make sense to anybody and it's only after a long period of time where my users are like nope that's not what it is it's this other thing finally i'm like oh okay you're right um so i don't know if you could tease out um yeah like and and like for example if you were giving advice to someone else, like me for example, I'm working on a project, the messaging doesn't resonate with anyone. What advice would you give to find messaging that resonates faster? Yeah, definitely. So a couple of thoughts.

First is I do look back somewhat frustratingly at the first two and a half years of this journey, not just messaging-wise, but clarity in general. We've wasted a lot of time. But I think that's sort of, hindsight is 20-20, and that's sometimes a necessary part of the journey. And that things that are obvious, I mean, it's the weirdest thing because when I came up with the grid interface, it felt like I'd thought of it before.

It felt so familiar, but I'd never done it. It felt like it was sitting there the whole time. And then when we did the website builder repositioning, it felt the same way. It's very hard to describe. It almost felt like I discounted it as a possibility because it was so obvious and I had to give myself permission to go there. But I think that's the nature of good ideas. They're just, they are so obvious

and that's misleading. And I think that we need to like train ourselves to actually um take that obviousness not as um like a repellent but actually a magnet that's a good thing that means there's something there like we always hear this idea of like um you know simplicity being this virtue and obviousness being good i think it's different in practice um and so just being familiar, like, I think as creators, we tend to want to do original things. And so there's an aversion to glomming on to any existing concept or metaphor. And I think we got to fight against that. The other thing I'll say in terms of messaging specifically is that I think one of the problems that all creators face, and especially product makers, and I'm guilty of this,

is sort of an egocentricity. Like we look at the world and we think that people care as much about us and the things that we build as we do. But the truth is that's not true at all. People have their own lives that they care about. And more often than not, you are not a part of that, nor is your product. And so I think that we are tempted to, because we really care about this stuff, think about it in really grandiose terms and at the expense of explaining in a way that's really obvious, really prosaic, really pedestrian.

And what I've come to realize is that if you're trying to do anything really interesting or new, what you want to actually do is wrap it in the most familiar thing you can, the most familiar form, the most familiar terminology. without people even really knowing it because it's really hard. It's really damn hard to get something new into the world to be adopted. So you want to use every unfair advantage that you could think of to bring something into the culture. And so for us, we had become enamored with this idea of creating our own web that had the values that we liked and all this stuff. And so we created, you know, this notion of a verse and, you know, it sort of, it was like 100% of what we'd want.

But we realized that actually, if we say, actually, you're making a website and Universe is a mobile first website builder we get like instantaneously someone understands like exactly what we're doing you know no one understood what we were doing before like no one had any my mom couldn't understand what i was doing like and as soon as it said it's a way to make a website on your phone, she got it. And even though that didn't capture this whole idea about, like, HyperCard and creating a democratic web, it didn't matter because, like, she gets it now, right? And I think you need to create, like, an interface, so to speak,

from a messaging perspective that is super, super familiar. And so I don't know if you've heard of this, but there's an industrial designer. Hold on, let me pull it up. So there was an industrial designer named uh raymond lowey uh he was at the time like he was he died in 1986 was born in 1893 he was um like the most famous he was a johnny ive of the the 50s uh so he designed planes and he like designed air force one and um and like trains and all kinds of stuff. And he had this principle and he called it MAYA, M-A-Y-A,

Most Advanced Yet Acceptable, in which he said you want to wrap new technology into familiar packages. And so you want to make the new familiar and you want to make the familiar new. So if you're making a bag of potato chips, you want to make that bag of potato chips seem like they're from outer space. Like they're so cool and new.

And if you're making a rocket chip, you want to make that seem like a bag of potato chips. Like that sort of idea. Yeah, well, to me, that almost feels similar that's sort of the idea yeah well i i to me that almost feels similar to the idea of the zone of proximal development for a child but more for like society so there's like you want to just go a little bit beyond the familiar for people because that's all that people can understand just like the tiniest bit beyond. Right.

Fascinating. Exactly. Yeah. Definitely, you know, in reflecting on your transition to my own work, as we talked the last time you and I talked, you were giving me the advice that I think you, I'm projecting now, but that I almost feel like you wish you could have given yourself, you know, two and a half years ago, which is like, kind of get stuff out there, build real products, talk to real customers, as opposed to staying in this abstract land of, of vision. So yeah, it's inspiring to hear how much more success you've had. I feel the visceral upward spiral that you've landed yourself on

and this incredibly long backlog of products that customers really, really want to use. It feels vibrated, the vibrations, I's if it like i it feels vibrated like you know the vibrations i can feel it so it's empowering so like you know thanks and and we're still like really like i said at the foot of the mountain and we have a long way to go but uh but it feels like uh like an optimization problem like you you feel like you're on a mountain and like you take a step and you're like yep this, this was higher than I was a second ago. Like it feels like you're climbing as opposed to just walking on flat ground. That is correct. Um, one thing I'll add though,

is that, look, I do think there's value in vision. Um, and, and so I want to make that clear, like knowing what you care about knowing your values um knowing your where you want to go is really important it's you know the that's the stuff that makes changes the world um i just think that it needs to be in balance with doing um and that uh a lot of the vision stuff is actually doesn't need to be that specific. It could be more of a general orientation. That's why I like to think about like values as opposed to like a specific vision per se. But I actually think like essential planning,

like a thesis is not the way that actually things change in the world. I think in academia that suffices, but if you actually want to instantiate change in the world, you have to engage in the world i think in academia that suffices but if you actually want to instantiate change in the world you have to engage with the world yeah yeah yeah that's that's my yeah yeah well when you when you engage the metaphor of like the central planner you know brings me to like the vision of robert moses the um yeah exactly it's robert moses versus jacob jane jacob yeah uh and then that's like a whole a whole separate debate um fascinating sometimes you need a by the way sometimes you need a moses character

right like if people like to give moses a hard time like and I definitely would choose Jane over him but um you know like New York I think there were some crazy amount of bridges that he built oh yeah and the city's better for them so like you know it's not one it's not just one or the other yeah yeah well it part of um I think to to kind of dig deeper into why I'm in this, in my current phase of vision and principles and all that kind of stuff now is because the more I read Brett Victor and Alan Kay. So, you know, my first blush at this was I watched a few of their essays and was like, wow, this is inspirational. Let me go build some stuff. And I built a bunch of stuff and I felt like it,

I wasn't that excited about it. So I went back and read my original inspirations. And the more I read them and listened to them, the more they like convinced me how important it was really like know your stuff. Like a quote that to me like epitomizes it is they say that like the internet, for example, like the core technologies of the internet, like the plumbing stuff were designed by like real experts, but then the web itself, like HTML, CSS, and JavaScript, like the work that Netscape did, the work that people at InterExplorer,

like that was more like amateurs that were just kind of hacking it together. And I think like Alan Kay has a lot of regret for his ideas that were kind of unfinished and hacked together that like spread so rapidly throughout the world. He kind of wishes that he didn't build them or he didn't release the code. He kind of like perfected it more before he released it. So I feel like I'm almost infected by that, that reluctance that, you know, to, to, to make things worse, uh, to, to like, it seems, it seems like Brett and Alan,

like they feel like the worst possible thing you could do is design something that the worst is better thing that designed something that's good enough to spread like wildfire and, but, but, you know, terrible enough to be worse than if you hadn't created it in the first place. So, um, I'm not arguing for shipping bad stuff, though. I just want to be clear. I'm of the less but better camp.

I'm just saying that it shouldn't be complete and that I don't think things are... Look, there are certain projects where you're shipping it out into the world and it's never going to change and so you want it to be finished, so to speak. But my point is that I think reducing scope such that the thing is as small as it needs to be, but high quality and not just high quality, representative of your values is critical. Yeah. Well, and it's so clear how now that you have this mandate

to make the easiest experience of building websites, making the experience of building a website as easy as possible. It's like so clear that like I know where you're headed. I can communicate to others

where you're headed. You can communicate. It's like very, very obvious. So yeah, that's the power of scope. So you've totally sold me on that um so anyways uh let's see i i have one or two more questions for you so i'd be curious to hear about what's on the horizon for universe now what's like the next target or milestone or like you know how do you conceive of the next 12 months or so how do you think about the future? Yeah.

A couple things. So we are continuing to iterate the product. We continue to allow you to make more powerful sites and more beautiful sites. So sites that do more things and sites that look better. Now, right now, sites of the universe are getting to the point where they're actually pretty competitive with what you could do with a desktop builder or, you know, come up with on your own. But I want to get to a point where you can do things with universe that you can't do with other tools. And you can't do them with other tools because they're not mobile first.

And universe is. So I want to really lean into the things that mobile is and not bound by the keyboard and mouse as an input. So you can expect to see our sites getting better. And sort of that's one thing. Second thing there is specifically around commerce. So a lot of people use Universe for their businesses. We handle these well, but we don't allow you to sell things or do anything really related to growing your business that will change um we'll support that and um we'll start to do a lot of the things like being able to to manage your site in a way that right now it's very simplistic you have one site you can you can't even change the title of it on

the web like the metadata um so we're gonna allow you to do stuff like that and make it easier to make sites that have multiple pages and things like that. But really leaning into like how to make sites that are more beautiful, more interesting, more powerful and expanding the portfolio of tools. The other thing that we're really interested in is opening up the universe block platform. So right now we offer I think seven blocks. It's like video, map, text, and each block has its own palette. So if you're in a text block then you have control over the font and the size and the positioning of the text and the color and things like that. If you're in the map block you can choose a style of map or color and things like that. If you're in the map block, you can choose the style of map or the address, things like

that. We recently introduced a code block, which is sort of an escape hatch for hackers. So you can go and write your own block and it will render as a web view within a block. And it's really cool because it opens up what you could do with the universe. So you can embed stuff, but you could also just write a you can write a ver a block from scratch um and we actually made a which i can link to we made a um a tool uh a desktop tool that allows you to write uh blocks and html and javascript and actually see how they how they render in our grid since a github project I'll share it but long term we

want anybody to be able to make native blocks which is to say that imagine you're a developer and you want to create a let's say a github block so you could wire up the github API create a visual representation's say, a GitHub block. So you could wire up the GitHub API, create a visual representation of what you want the blog to do. So let's say display the five most recent projects that you've contributed to. And then for the user, the creator of the site, they would just, in a native way, enter their GitHub username and it would pull, you know, sort of then inject that into your code. And so we want to open up the creation tool itself to other developers.

And, you know, the code block is the first step there. And if people are interested, they can go check that out and start writing blocks against that. They can actually even submit their code to us and we will, if it's good, approve it. And then any other user on the platform can use that code. Long-term, the goal is not for the code to be the interface, but rather for us to build out this more native platform. So those are some of the areas that we're focused on. We're also focused on growth, on getting in front of more people, and on allowing people to get their site in front of more people.

Because people make something awesome. They want an audience. Yeah. Totally. I imagine analytics, as we've talked about, is probably one of the biggest requests you audience. Yeah. Totally. I imagine analytics, as we've talked about, is probably one of the biggest requests you get. Yeah. Yeah, and when you were talking about enabling commerce,

I was almost thinking, Squarespace also allows commerce, but I was thinking that you could almost be a mobile for Shopify. I'm imagining some designer who, instead of wanting to sell their things on Etsy, they could just take pictures on their phone, apply filters, and then just put prices on them and then just sell things all without leaving their phone. They don't even need a computer. That's very futuristic

to have a whole business that you run from your cell phone. It's crazy. I also say it's already happening. It's just the tools haven't caught up. That's why we, I mean, one of our top requests is commerce. And the other thing I'm going to say is like, one of the things that we realized is that, you know, the website builder space is very crowded. There are Squarespace, there's Wix, there's Weebly,

there's Strikingly, there's all of these tools. And it's a very big business now. Wix is a publicly traded $3.5 billion company. But there are virtually no companies that started on mobile. Now, that's a nuanced point. Weebly, for example, has a mobile app. But it's a derivative of their desktop app. So it's moored in the conventions and the paradigms of desktop.

We're the only ones who started on the phone, and that gives us a different perspective. So that same lens applies to commerce. Shopify has a mobile app. We look at it as starting with a mobile app, and that sort of reduces the scope. Totally. I think it's pretty clear when you use your app that it was built for the mobile phone

and it was built to build mobile websites. So I think that comes across through design. You don't even have to tell users, like, you know, by the way, we were on the phone first. It's like, it's just, it's clear because the scope is reduced and this is your entire focus. So you're able to make a really high quality product. I was emailing with someone who used to work at Squarespace, was curious about their mobile efforts. And they told me that, you know, they tried an app that allowed you to make a cover page for your eventual real

quote-unquote website and it it for them was really like a lead generation tool so they can drive download like drive signups for their their main product um but sort of target people on a phone and didn't work and and he had said the reason why it didn't work was because at some point you know if you want to make a change to your site, it's, quote, unquote, easier to just open your laptop and make a change. Yeah, that makes sense. And, you know, look, that makes sense if you're like – Well, what I'm saying is it makes sense given that their mobile interface was so bad that, like like given that their interface was built for desktop like you know people just wanted to use the desktop interface because it was the better one

exactly and but but the orientation is a desktop first orientation and that is sort of missing the point the point is that um the doing it on your on your computer is the last thing that most of our customers would think of wanting to do. So here, I have one last question for you. A lot of people in this world of building the future of human-computer interactions are concerned about, including myself, are concerned about the long-term financial sustainability of our work. So I know that you've raised money and now you are making some amount of money from your customers through various services you charge for. So I'd be curious to hear about how you think about that. Are you thinking about continuing to raise VC money? Are you focused on revenue in the short term? How do you think about continuing to raise VC money? Are you focused on revenue in the short term?

How do you think about that? Yeah. So we are VC-backed currently, and we're likely to continue on that trajectory. With that said, our business model is subscription based. And so we will offer different products. So right now, to create a site, it's free, but you can subscribe to a custom domain. And we're adding an additional subscription today, actually. And we'll continue to roll out new subscriptions that over time will get us to a place where we're sustainable from a cash position. The big point there, though, is that we're building a business and a business model that is aligned with our customers. So we make money when our customers

are happy and we are not at odds with them. So, for example, like I was saying earlier about when we were building a social network, that felt start to feel bad because really our incentives as a company were not aligned with our users. Right now, we're really aligned. And companies like Wix and Squarespace have demonstrated that you can build a very, very large business with this model in our space. So that's really promising.

And the thing I want to say is that VC, there are bad actors. I've had my share of bad experiences, but VC is not categorically, it doesn't have a moral stance categorically. It's not bad or good. It's a tool. It's a tool like any technology. And it's really good at allowing you to swing big and do exciting stuff and start a high-growth business. And these are all amazing things. They're amazing innovations. And what I would say is that's more insidious,

that's more problematic, is people choosing business models that are at odds with the people who they serve. Now, VCs will often encourage or be supportive of that because, again, it's amoral. And I think that's the thing to watch out for. But this is often within your control. I think we don't often think about a company's business model when we think about its role in the world. And I think that's wrong. I think that we need to put as much thought and care into the business model as we do into its design, for example, its user interface, its technology. Yeah, that totally makes sense.

I think from that perspective, the VC tool seems like you're using it right. Yeah, like, I mean, we'll see. I'll tell you this. We're not in a position right now where we're cash flow positive. So we're not profitable yet and we have to get to that point. But, you know, I'm confident that if we continue to do a good job, we will be in a place that's sustainable, which is cool, right? We're properly incentivized as entrepreneurs, right? And too often, it's the case that to actually be financially successful, you have to do things that are not in the best

interest of your users. Like I think Facebook, for example, is a very high quality sort of executing machine. But I would argue that more often than not, their incentives are at odds with what's good for their users. And the more successful they are, the worse off in some ways their users are.

And that's just bad. Like I don't think that is a good thing and i don't think that like we don't we don't really need more of those in the world like i think it's up to us to use like i think if you're an idealist uh it's really important to figure out how to create a real world vehicle or structure that um can over time bring positive change into the world. And I think the best way to do that is to design a business that has the potential to be successful when its customers and users are successful. Yeah, I think that's very noble of you.

And I like the distinction that it isn't venture capital that is immoral. It's that sometimes it's often that I think it's that people conflate those things because venture capitalism, because people often use raising VC money to build a high growth, free viral product that doesn't earn revenue for a long time. And then eventually they cash in on the attention of their users to advertisers sometime down the line.

So VC is used to defer revenue until you're able to cash in the attention in this attention economy, which I think you're correctly pointing out as being immoral because people are stealing people's attention through these very like nefarious mind hacking kind of notifications, personifications. Yeah, isn't good for people. So, but yes, you, on the other hand, are using venture capital dollars to do really amazing fundamental user interface research. And then once you've figured all of these key things out, you're now aligning your incentives correctly where your user and your customer are the same person. And they pay you when they're happy, as opposed to selling the eyeballs of one person to the use of another person. Yeah, look, I think we have a real challenge, though.

I don't want to – like we're not there yet. Like we're not financially sustainable. And that, if we're going to be around for a long time, has to change. So this is a challenge. But I think part of starting a values-driven company is sort of not wavering on that. I think one of the things I've learned is how important the business model is. It really matters. Yeah. Well, anyways, on behalf of all of us, thank you for starting a values-driven company like this.

And thanks for the noble task of trying to find this global maxima in terms of human-computer interfaces. I think we'll all be better for it. Thanks. I mean, we're going to need a lot of help. So I appreciate any ideas people have. They want to check out the product. If you guys check out the product and you have some suggestions, you can email me, joseph at onuniverse.com. I would love to hear your feedback.

And, again, like I said, we're just getting started here. Cool. Yeah. I'm excited to see what day two and day three looks like. It's a very promising beginning. So hopefully, maybe I'll have you back on here in a bit and we'll get to hear the updates.