At its best, art is transcendent and sublime. However, few artists can regularly create works that evoke feelings of awe and wonder on that level in its viewers. Even fewer artists have managed to do this by writing code on a computer, a tool which many consider to be at odds with natural beauty, and sometimes even nature itself.
Among generative artists, Jared Tarbell is singular in his ability to continuously create works that rival nature in both their complexity and grace. He has been doing so for decades. Tarbell’s work is beyond timeless. The images he created in the early 2000s still feel like they are light years ahead of where we are with generative art today.
It was Tarbell’s work that turned me from a bratty young studio art major who dismissed all digital art into a lifelong advocate for generative art. I have written about Jared’s work for years. I’ve also included his pieces in exhibitions that I’ve curated around the world. However, I never had the opportunity to speak with Jared personally until now.
It turns out Jared does not do very many interviews, which made this an especially great honor for me. Early in the interview, I discovered that, like myself, Jared was raised Mormon and has since left the faith. While I left that out of the interview, it gave me important insight. I see Jared’s work as the reconciliation of the intrinsic spirituality of the church and the inherent logic of computer science. This is a childhood experience that I personally know very well. I believe that through his art — and particularly, his embrace of emergence — Jared seeks to find harmony between the seemingly oppositional forces of the spiritual and the rational.
Our interview comes in the context of the show “The Game of Life - Emergence in Generative Art,” currently running online at the Kate Vass Galerie. The show explores the concept that very simple systems or sets of rules can explain much of the complex behavior we see in the world around us. This idea of ‘complexity evolving from simplicity’ is core to Jared’s work.
I am thrilled to contribute to the relatively small amount of literature on this incredibly important but private artist. Enjoy!
Interview
Jason Bailey: Jared, I consider myself the president of your unofficial fan club. I’ve looked for recent interviews on you and your work for years and not much has come up.
Jared Tarbell: Let me tell you, I really admire you. I think I’ve read most of everything you’ve written. And, you know, I don’t do interviews, and I’ve got a lot of requests. Most people want to talk about Etsy and how to create a billion-dollar startup. [Interviewer’s note: Tarbell co-founded the online retailer Etsy in 2005.] So doing this interview is a rare thing, and I’m really happy that I get to talk to you.
That’s maybe why there isn’t too much history or too many recent interviews on me, because I’ve realized with a little bit of public attention that I don’t enjoy it. I really do like being kind of private with my life and my work, because I don’t really don’t take myself that seriously. It’s really your perspective and your insights that’s compelled me to do this interview.
Jason Bailey: Wow, thanks, that really means a lot to me.
I’d like to start by asking a question I ask a lot of generative artists. Did you come from math to art or from art to math? Or were they parallel interests that kind of drove each other?
Jared Tarbell, Bubble Chamber, 2003, Processing sketch
Jared: Great question. I was encouraged to make art in the home by my mother quite a bit. But my background in college was mathematics and computer science, so I really viewed the world that way. I also wasn’t afraid to try to express myself or try to find beauty in computation. I tend to view the world computationally — the idea that this world is a very complex system and that it’s hard to try to identify the rules that are driving it. It’s just some kind of natural phenomenon that we don’t understand.
But in reality, you can break the world down into these very simple rules that interact on orders of magnitude beyond what we’re capable of imagining. It’s a result of this phenomenon, this emergent phenomenon. I don’t know if it describes everything. In fact, I kind of don’t believe it does. I think there still are some very spiritual, mysterious elements to the universe. But certainly, thinking about problems computationally, you can get pretty far in understanding what’s actually happening.
Jason: That’s really insightful. So did your interests in programming start in college?
Jared: I was programming at the age of 14. My dad brought home a computer from the labs. It was, like, a Sanyo 8088. And more importantly, he taught me how to program in BASIC really simple little things such as number guessing games, like the game Battleship where you pick a number and it’s higher or lower. I eventually moved on to Pascal and created a two-player version of Tetris. You can actually still find it online. I think it’s called Atomic Tetris.
Jared Tarbell, Atomic Tetris, 1991
So I was really into computers, and I knew that I wanted to study computers in college. In fact, I sold one of my first games — that Atomic Tetris — to a software bundler. They sold software bundled with new computers. It was Packard Bell at Montgomery Ward. You would get this collection of games. It paid me a few thousand dollars, which doesn’t seem like a lot, but at the time, it was huge. That’s like a few thousand dollars before college. It helped me pay for my tuition the first couple of years.
I studied computer science right away, eventually learned all the languages they force you to learn in college — C, Pascal, and Miranda and COBAL. But there was no web work.
I really wasn’t working as an artist. I was interested in making video games. That’s what the gateway question was for me. How do I create video games? Because they were just so fun at the time!
Then getting out of college, I realized the web was really coming up, and I really wanted to make a web page. Just something, but I didn’t understand what it was yet. And I really wasn’t creating these algorithms — like, these drawing machines that I think I’ve become well known for — until later when I discovered Edward Tufte.
I was a computer programmer like most computer programmers, making just the ugliest websites you can really imagine. Functionally perfect, working and taking your data, but just really hard to use and ugly. And one of my early employers sent me to an Edward Tufte conference and said, “Jared, just go spend the day listening to this guy.” You’re familiar with him?
Jason: Yes! I never thought about it, but I can see how your color palette may have been influenced by Tufte?
Jared: Yes. Absolutely. And so many of my design aesthetics came from him. He breaks down design with visual rules. My very analytical mind at first says, well, okay, if I’m making a grid, I don’t need to draw the grid. I just need to draw the elements. And instantly, it was like, wow, that actually looks really good.
Jared Tarbell, Sand Dollar, 2004, Processing sketch
Jared: Another particularly powerful lesson I learned from Tufte was small multiples, which applies to generative systems. You can build this machine that generates an infinite number of forms, but all very similar forms. But how do you show what the machine is doing to people? The way to do that is by just laying small multiples right next to each other. When the outputs of the machine are all next to each other, you get an idea that this is a system — this isn’t a single image. I think it was after I applied Tufte’s rules and I started writing these programs that were actually creating beautiful output, that’s when I thought, well, maybe there’s room for art here.
Also, I was really harsh on artists. I thought, oh, look at these guys, these artists are so selfish. They’re just doing what they want to do and expressing their view of the world, and really not contributing to the world at large. I thought, what good is this? These artists are just indulging themselves. So I really stayed away from the realm of art, and I never wanted to be called an artist for a long time because I had a more functional view on how I wanted to contribute. I wanted to build a tower or a spacecraft or something of this nature. I didn’t want to inspire or ask questions about the nature of the universe. Which, of course, is all I do now.
Jason: [laughs] So what changed you from your artist-hating younger self?
Jared: I think seeing the impact that art can have on a person, or even have on a field. I applied some of these artistic concepts to some of my early work on Etsy. We needed a tool to shop, and I thought, well, let’s apply some of these artistic values to this tool. And seeing just how powerful that was — you know, it helps sell items, one very real thing. So I thought maybe it’s okay, because here’s actually some function that comes from it.
Also, finding artists and being influenced by artists myself. I realized that reading or enjoying music or looking at paintings changed my life. I saw the power. I thought, okay, maybe this isn’t meaningless. Maybe this isn’t just selfish work. Maybe I actually can contribute in this way.
Jason: Did that change happen to you in like the early 2000s?
Levitated.net 2000-2004, open source examples of drawing machines by Jared Tarbell done in Flash
Jared: So levitated.net was 2000 to 2004, maybe, where I was coming up with daily Flash sketches.
And then after that was Complexification.net, where I started to get into complexity theory. I would say that transition happened in 2004. That’s when I started working as an artist, and even selling my work.
I would create prints of my work and sell them, so certainly I was trying to make a living as an artist, which we all know is really difficult. I wrote my own shopping cart, which took three months of grueling work that I didn’t enjoy, but I knew I needed to do it because it’s work I needed to do to sell my work. And it was, in fact, through that experience that I attracted or met my company-founder at Etsy, Rob Kalin. He found my work through Complexification and invited me to work on Etsy with him. So 2004 would be the transition.
Complexification.net, 2004-2005, open source examples of drawing machines by Jared Tarbell done in Processing
Jason: Were you able to find time to make art in the startup days of Etsy?
Jared: No. It went away almost completely. Etsy was definitely a super-full-time gig. Wake up, before I even got dressed I would check some of the site’s stats, respond to people in the forums, and basically work on the site throughout the entire day. In the beginning, I remember sometimes I would have conflicts with some of the other engineers or my other co-founders, Haim Schoppik and Chris Maguire. We’d get into arguments about the best practice or best way forward, and I would also fall back on my Alexa page ranking of Complexification. It was ranked higher than Etsy at that point. So I was thinking, I’ve got this art website that’s outperforming Etsy. I could just go back to this. You guys better get in line. [laughs]
Jason: [laughs]
Jared: But, you know, we all worked together really well. Very quickly, after about the second year, Etsy eclipsed Complexification in terms of the number of people that were interacting with it, the number of people we were reaching, and the amount of time I was spending on it. Complexification just kind of dies around 2005. There’s actually a sketch called Colony that I was working on and didn’t quite finish, and I put this little tag on it that said “soon.” And that’s still there to this day. I’ve never actually returned to that site to finish some of those sketches. It wasn’t until I left Etsy in 2011 that I was really able to return to my art.
Jason: Then, in 2011, when you come back to your art, is that when you build out the Levitated workshop? I think there was sort of a like a laser cutter time period?
Example of work created at the Levitated Toy Factory in Albequerque, New Mexico
Jared: I moved back to Albequerque. I was living in the country, pretty remote. I bought this old building and renovated it. I was really interested in this laser cutter that I bought at the time. It’s part of my trying to get my work off of the screen and into some more tangible, physical form. I thought I’d wrap a company around it, like Lego, where we would create some kind of a building system that was open-ended and used a lot of these algorithmic techniques.
I called that site Levitated.guru, so I used the new top-level domain “dot guru.” It just kind of froze — that’s what I tend to do. I just freeze the sites that I’ve worked on in the past. So Levitated.net today looks a lot like it did back in 2000. Same with Complexification.
And the Levitated Toy Factory lasted a couple of years, about 18 months. I don’t know if you want to get into that. But running a company is really difficult. I found that what I was doing every day was just managing the books and the people, and I really didn’t have any time to do any of the creative work that I would like. So with the birth of my second daughter, I used that as a great excuse to just close the door to the Toy Factory and devote myself really to my family at first, and then slowly, as the babies grow and you have a little more time to yourself, that’s when I’ve started to get back into the artwork.
Jared Tarbell, Wilson Maze, 2019, Baltic birch, acrylic paint
Jason: You have a new site that I recently discovered. I was pretty psyched. It was like stumbling onto a goldmine. I was like, woah, Jared’s making work again! I should know the URL, but the new site is —
Jared: — infinite.center.
Jason: How are you looking at that site or your work now that you’re kind of going back to code-based work after all the experiences you’ve had with the Toy Company and Etsy? Being back with your new life experiences, how has it been?
Jared Tarbell, Penrose Substitution. 2020, Processing sketch
Jared: It’s been great. I’ve realized that I’m in this position where I can do almost anything I want to do. But what I really want to do is just write code and share it with people. So that’s what I’m doing. The sketches that are on infinite.center are mostly drawing machines. And I haven’t really found a way to incorporate some of the laser cutting work that I’ve done. It’s hard to share those projects. I almost want to keep it pure code versus code and assembly, you know, which is the laser work. And I haven’t figured out a way to share that work.
So this is how I spend my day. I think of an algorithm and I spend a couple of days working on it and try to get it to a point aesthetically where I’m relatively happy with it, and I post that. I’ve also committed myself to open-sourcing every project henceforth until my death, I hope. Which is difficult to do. It’s actually really easy to write a program and share some images or video of it, but to actually go and open the source code, you know, it’s a little terrifying to do that because you’re not secure — I’m not secure, necessarily, with everything I write. I’m not the best programmer. And it’s also a lot of technical work to get everything just right, upload it to GitHub, or even to find the platform that’s necessary to share the work. That’s what I’m trying to do now.
Jason: I want to ask you follow-up questions on your process. I think Casey Reas and a few other artists have mentioned that the algorithm or the core structure of what they make actually goes relatively quickly. However, it’s the tinkering, tweaking, and the adjustments that can take days, weeks, or months. I wonder is that true for you?
Jared: Yeah. I love that. Casey’s one of my mentors. I really admire him. I’m so thankful to him and Ben Fry for creating the Processing language, which is primarily what I work with now.
Jared Tarbell, Vesica Piscis: You, Me, We, 2020, Processing sketch
Jared: I would agree with what Casey said. The core functionality of an algorithm comes together really quickly if you’re lucky, within a matter of days, usually. But then finding the expression or what you want to present with that algorithm to kind of explain the process, that takes a lot more time.
Part of the problem is, there are so many choices — because you’ve built this parameterized system which is multivariate there’s anywhere from three to a hundred different variables that you can tweak to affect the output of this program. And so one program really can express itself in a dozen or more different ways, each with something that is aesthetically unique. So finding the right one or finding the particular parameter set that you want to display or share with the world, that takes a long time. It’s hard because sometimes you have to abandon a pretty cool idea or a pretty cool expression to move into what you think might be even more interesting. That’s difficult.
I would actually really enjoy hearing Casey talk about that.
Jason: Do you think people need to have a certain level of programming literacy to fully appreciate generative art?
Jared Tarbell, Buddhabrot, originally written in 2004 (took six days to render).
This render was created in 2020 in 15 minutes.
Jared: Programming, you can break it down into two categories. There’s the semantics of the program — what are you actually trying to do logically — and then the syntax of the program — how are the characters arranged, how does everything flow together.
Syntax I don’t think is very important for anyone to understand, unless they want to have a career in computer programming or if they want to do it themselves. But the semantics, understanding the logic, I think that’s important because it gives a person a lot of insight into what’s going on around them.
So let’s look at the algorithm Substrate. You can understand the semantics of that program very easily. Like, this is the rule, it’s what’s happening. There’s a line that goes straight until it hits another line or the edge of the screen, and then new lines form at right angles to existing lines. That’s all you need to know. There are just two rules. And then watching the whole system unfold, I think it gives you an idea of the potential depth. It’s just this really simple thing that’s operating, but look at the complexity of the results that are being generated.
The best algorithms that I’ve created are the ones where the semantics are easily understood or described. Then you have a deeper appreciation for the results of the algorithm, the resulting image.
Jared Tarbell, Substrate, 2003, Processing sketch
Jason: I’m glad you mentioned Substrate. It’s one of my favorites. It reminds me of the cracks in the desert. Was that your inspiration?
Jared: Substrate is one of my most popular algorithms. I was sitting in a coffee shop in Santa Fe and there was a sticker in the window blasted by the sun. It separated into these cracks. I saw the algorithm there and then and wrote it in that coffee shop. The desert sun is brutal and did that.
Jason: Wow, I love that story! I’ll never look at Substrate the same.
Given that you look at the world sort of computationally and that you have, I think, a strong understanding of what we’re calling the “semantics,” are there a fair amount of surprises as you’re creating your work when you execute your code? Do you have a blueprint in your head that you are executing on or is it more non-linear and improvisational than that?
Jared: That’s a great question. I actually look for the surprises because I’ve been programming the computer for so long, it can get really boring. Like the number-sorting algorithm. There’s an algorithm that’s going to sort numbers. There are a hundred algorithms that sort numbers. But in the end, the numbers are always just in order. That’s not very exciting.
And so when I discovered emergence and some of these generative algorithms that produce surprising results, that got me really excited. Finally I was using this incredible machine with these ideas, and I don’t exactly know what the outcome’s going to be. I have an idea, but it’s not until you run the program, and then it unfolds before you. It’s like, wow, this is really interesting. So I search out the surprises.
I’ve found some particular classes of algorithms that generate those surprises quite easily. They’re usually systems that involve many simple elements that are allowed to somehow interact with each other. And then you just let the system unfold. You always have an idea of what you’re trying to do, but it isn’t until you run the program and see it unfold that you see the results.
Jared Tarbell, Spiritual Noise Machine, 2019, Processing sketch
Jason: I’m going to have to wind down in a little bit, thank you for being so generous with your time!
I always like to ask, are there some things that I haven’t asked about that maybe you would want to share or that you think folks would find interesting that maybe people don’t know about you?
Jared: Well, not so much about me, but a lot of people ask me, “How do you become a good programmer and create meaningful work?” My advice is always this. Just program something every day, even if it’s something simple. You can just do a little bit every day, even just 15 minutes. You’d be surprised how good of a programmer you are by the end of the first month of doing something like that.
Jason: That’s great advice. I think I need to take it. I got pretty good in grad school at Processing, and I thought, well, it’ll be like a bicycle. I’ll get back on it and it’ll be completely where I was 12, 13 years ago. But it’s not a bicycle [laughs].
Jared: [laughs] Don’t give up.
Jason: I won’t. You’re my inspiration. Now that you’re making work again, I know that I have to make work again. I just won’t compare myself to you because then I’ll feel bad. [laughs]
Jared: [laughs] Thanks, Jason. I would be happy to talk to you anytime and in any context. I hope we get to meet someday in person after this whole virus thing is over.
Jason: Yeah, I’d love to get out your way and see New Mexico again. My wife and I love it. And if you’re ever in the area, it would be awesome to hang out. Who knows? I think both of us are a little bit more on the introverted side. But if a cool enough opportunity comes up to do a panel somewhere, maybe we'll sneak out and do the panel and then get the hell out of there and have a fun adventure somewhere.
Thanks for your time, Jared. It was really exciting.
Learn more about Jared’s work and the exhibition “The Game of Life -Emergence In Generative Art” at the Kate Vass Galerie Website.