Interview by Miguel Penabella
Originally conducted December 12, 2019
Edits and introduction written September 1, 2020
When I caught up with artist, professor, and game maker Everest Pipkin late last year, they were wrapping up a semester of teaching but still finding new creative outlets for their talents and expertise. Outside of the college classroom, Pipkin has remained diligent in compiling accessible resources for artists and students alike, penning guides and other scholarly essays available online. These written pieces comprise just a fraction of Pipkin’s diverse body of multidisciplinary work, and their career resists easy categorization. Pipkin has produced digital work that pulls from large data sets and online archives, from neural network generated zines to Twitterbots that generate renditions and names of imaginary moths, to Default Filename TV, a channel that queues YouTube videos with unedited filenames like “MOV 5681” or “IMG_6397,” showcasing amateur videos of pets or teenagers goofing off. These various projects are all part of Everest Pipkin’s goal of fostering a more “intimate internet,” one that values individual creation over corporate control.
Talking about the relationship between human beings and the platforms with which we choose to engage on a daily basis, Everest Pipkin emphasizes the potential that these spaces can be ones of intimacy. If computers and other devices serve as our primary means of connecting with other people, these technologies can be considered a home or a community, deserving of our care. These ideas are partly why the ecological world factors so prominently in their game work, compelling us to break down the barriers between lines of code and the natural landscapes in which we live and let flourish. The Worm Room, featured in the XXI Triennale International Exhibition in Milan alongside other experimental games, takes players through endlessly generated glass greenhouses along a garden path sourced from the Biodiversity Heritage Library’s botanical illustrations. Untitled (the flower game), developed with frequent collaborator Loren Schmidt under the moniker Withering Systems, draws from the bullet hell genre to create kaleidoscopic floral patterns that slice, spiral, and spread ever outward. Other games, like mirror lake or inflorescence.city, defy traditional norms of videogames with markedly stripped down and minimal forms of interaction. Pipkin coolly discusses their interest in frustrating the definition of games, shrugging it off with a smile and pinpointing an interest in the poeticism of slowness and other decorative possibilities to make things feel like home.
Sitting down for an hourlong talk with Pipkin cannot even begin to cover their wealth of knowledge and perspectives, though they generously share ideas and memories from the classroom to the wider web. We reminisce about intimate corners of the internet from Neopets to GeoCities, growing up playing HyperCard games on the Macintosh Classic, and speedrunning Pokémon Blue before the batteries drained in the night. Pipkin and I discuss the appeal of procedural generation and bots, and the importance of curation and discovering sources you care about. They detail the thought process behind games like The Worm Room and inflorescence.city, and consider the broader connections between computational work and the ecological world at large, embracing a more inclusive, intimate, and gentle internet. Below is our interview (conducted December 12, 2019), which has been condensed and edited for clarity.
I’ve noticed very recently that you’ve put together some very helpful guides online. One of them is a guide for navigating the Unity 3D software, written in very accessible language. The other is a guide to resources and funding for artists on the Internet. What motivated you to write these guides?
It’s a boring answer: I’m teaching this year! For the first time, I have a full-time teaching load. I’ve definitely been a teaching assistant and have done workshops before, but this is the first time I’ve been a full-time faculty member. As I’m sure you know, being from academia and probably having students of your own, that these positions can be a reorientation of practice towards usefulness. There is something about being in a classroom with folks actively learning from you that puts into sharp relief all of the skills you’ve picked up and internalized over time. These aren’t necessarily skills that other people know yet, even if you feel that it’s common sense. This is maybe truer for the artist opportunities guide, which is not exactly institutional knowledge. There’s not a Word Doc anywhere; it’s the type of thing that you pull together individually as an artist to make your life possible. You figure out your own patterns, places, and strategies, and of course, that changes from person to person.
Talking to students who remind me of where I was at that age—not even ten years ago—it’s remarkable how much of that stuff I’ve picked up and turned into my to-do list for the day. Those are skills that you can teach people. It’s kind of a shame that they’re not a large part of arts education. I mean, they are a part of arts education, but I feel like it depends on what institution you’re at, what teachers you have, and the opportunities and funds that they’ve cobbled together for themselves. I’ve assembled those resources for myself forever, and I put them in a presentational format for students. Once I had it together for a classroom, I thought it should just be on the internet too, because if there’s anything worse than knowledge that doesn’t make it to the institution, it’s knowledge that gets to the institution and stays siloed. That’s maybe even worse. It’s been nice cleaning up some resources for a similar Doc for a p5.js. p5.js has a mellower documentation, so it’s maybe a little less essential. For me, as somebody who finds reading things in order to be useful, I think it could be handy for some. I’ve also got one on Bitsy Game Maker. Basically, what happened is I’m teaching, and a lot of my focus and energy that would normally be going into my arts practice is going towards thinking and writing about how to make those tools accessible.
“These positions can be a reorientation of practice towards usefulness.”
Can you tell us about your background and education? Where do you come from, and how did you get to where you are now?
I followed a pretty standard American academic track. I went to a decent public high school, and a pretty good state college, and then a private graduate school. You get lucky with all of those minimal state supports that are left. [laughs] I’ve been in and out of the educational institution for probably my whole life. I think it’s a space that is compelling to me despite its problems because it’s good to be in an environment of learning. It’s remarkable to be around people who are actively involved in their own practice and making things. Engaging with artists who have chosen that institutional or educational path as opposed to one in a gallery-focused space has been really lovely. They tend to be great people, and so those environments have been helpful for me. In some ways, I feel like I have had more experience teaching or collaborating in that way outside of the institution, in projects like the Museum of Human Achievement. I worked as a gallery coordinator for a couple of years, and I am still among the board. I was also a student and teacher at various points for the School for Poetic Computation in New York, and those types of non-hierarchical, non-accredited learning environments have been really rewarding for me. I’m trying to pull some of that ethos into my classroom.
What are some of the classes that you’re going to teach, or are currently teaching?
I just finished my semester today, which is why I was like, “Let’s talk today!” [laughs] It has been busy. In the fall, I taught a sort of Intro to Creative Coding class at Carnegie Mellon University. It’s an interesting class; it’s required for all art students, which is kind of rare. It’s an interesting crowd, mostly second-year undergrads who are nineteen or twenty years old, and the majority of them are not folks who consider computers to be a part of their practice. Some of them don’t have a history with working with computers at all. Some of them don’t have laptops. It is both a really fascinating challenge and a really inspiring course because you are teaching Unity, programming, computational tools, and asking questions like “What’s a website?” to folks who are painters, sculptors, and performance artists. The ways that they engage with that media is sometimes a little left field. It’s like, “Here’s an object. You know what an object is,” and they’re like, “What can an object be?” [laughs] So that class has been interesting and cool. We’ve covered p5.js and Unity, and I spent some time on open-source and accessible artist tools like Twine, Bitsy, and Nathalie Lawhead’s Electric Zine Maker. We talked about game poems and Fluxus, and the interactivity in that space of written instruction. We also made our own homepages, which was fun.
“If there’s anything worse than knowledge that doesn’t make it to the institution, it’s knowledge that gets to the institution and stays siloed.”
In my advanced class—which is an elective called Data Gardens—the students that end up there tend to be more focused on computation as the nexus of their work. It involves everything that comes bundled up with being an artist on the Internet, so topics like the ethics and morality of using big data and internet infrastructure in your work, the ecological cost of those things, the physical location of it, alternative networks, a decentralized web, web histories and potential web futures, surveillance, and data gathering. It’s that big, complicated field of “Oh no! We work with tech!” That class has been really lovely, and I am excited about everything that people have made and all of the resources that we’ve put together.
You mention that a lot of your students come from this wide variety of backgrounds, and you yourself work with a lot of different mediums like software, bots, books, drawings, zines, etc. But you also make games! What drew you into making games?
[pauses] What did draw me into making games?
What’s your earliest memory of videogames?
I had a fraught childhood relationship with videogames, which is that I loved them more than anything in my entire life. I coveted them but was not allowed to have them, which is maybe why I’m still so invested in that space. [pauses] I mean, that’s not entirely true. My lovely, very well-intentioned parents probably correctly were like, “No game consoles in this house. Limited screen time.” As a person who works primarily with computers, I was appalled. [laughs]
So many developers I’ve talked to have had the same story, where they weren’t allowed to play videogames as kids.
Yes, which is I think maybe part of the reason that you become fascinated with game worlds in your imagination. You spend all of your time pretending to be there, and then you can go and make them. When I was little, we had an already outdated Macintosh Classic with HyperCard installed. It lived in the living room, and my little sister and I were allowed to play with that. As far as I can remember, we had maybe six games including Cosmic Osmo and the Worlds Beyond the Mackerel and The Manhole, which are both exceptional old-school HyperCard games from Cyan, the same developers who did Myst. So, it’s a very interesting trajectory. Those games were released in 1989 and 1988. We were probably playing these games around 1996 or 1997, so the world had turned but we hadn’t necessarily caught up. We also had HyperCard, which is in its own capacity sort of a game development tool. Although I was a little bit young to be making complicated, cohesive worlds, I definitely had some sort of early exposure of like, “Okay, I’ve got thirty minutes of non-network screen time a day, and this one program. What can I produce in it?” I think HyperCard is still a bastion of really exciting software development for a non-coding-based user.
“It is both a really fascinating challenge and a really inspiring course because you are teaching Unity, programming, computational tools, and asking questions like ‘What’s a website?’ to folks who are painters, sculptors, and performance artists.”
After that, I have strong memories of borrowing my childhood friend’s older sister’s blue Game Boy with Pokémon Blue, which I was allowed to keep for a month because she was done with it, so long as I didn’t save over her file. And so, I sequestered it in my bedroom and hid it under my bed. I would go upstairs, get in bed, turn it on, and play until the batteries died every night. I never managed to beat the game in that five or six hours of wakefulness when I should have been sleeping, but I got pretty close. I spent an awful lot of time thinking about my speedrun strategy: “I’ve got this much time. I could get the Clefable. I know my favorite Pokémon is Pikachu, and there’s one in the forest.” So that sort of interface with that world. I also thought about the glitches that are predominant in the Pokémon games, which can help you once you get to Cinnabar Island. I think you basically win because you can get infinite rare candy and just trek on through. I streamlined my experience to be something that was still emotional and that I still cared for, but it was basically a speedrun strat so I could potentially beat the game before school or before the batteries died because I couldn’t save over the file. This was maybe seminal and important as a kind of weird gameplay thing. [laughs]
Later on, our library in our school got a network connection, and then finally, my house. At that point, I got on Neopets, which I think is a big touchstone for a lot of people my age. Neopets itself is a kind of remarkable and weird example of “pets culture” from the early aughts, and they also had homepages for your pet that you could code in HTML, which was really remarkable. I was a little too young for GeoCities. I didn’t have that type of other accessibility in my life of like, “Oh, I’m just writing code, and this stuff is happening on the screen.” So, I learned HTML and CSS through that, and then got on the Neopets forums where there was, as is wont to be, a rich culture of roleplay. That then kicked me over to the Avidgamers sort of trashy, text-based roleplay service, and at that point, I started running text-based roleplaying games which probably went on until I was fifteen or sixteen. That was my childhood of weird gaming experiences throughout that time, as I never had a console. I would go to all my friends’ houses and be like, “Thank you! Please!” Looking back on it, I think it’s the type of media that I still find compelling and effective, and I’m grateful that I grew up in a space that facilitated building digital worlds regardless.
“Okay, I’ve got thirty minutes of non-network screen time a day, and this one program. What can I produce in it?”
I want to talk about your game The Worm Room, which was created for the XXI Triennale International Exhibition in Milan, Italy, and later compiled for the Triennale Game Collection alongside games by Cardboard Computer, Tale of Tales, and others. How did you come to be involved with that exhibition?
They just sent me an email [laughs] which was very nice! I think [exhibition curator] Pietro Righi Riva had seen mirror lake, which had come out maybe a couple of months before we started talking. Because I went through an arts school education and had made computational-based work and drawings, I think maybe luckily, I missed the whole “What makes a game, a game?” discourse. I wasn’t around game spaces in 2012 or 2013 discussing playability; I was in art school, thinking about sculpture. Having found my way back to not just being interested in games as a medium but as an extension of my arts practice—probably around 2015 or 2016—I was making weird, lightly generative computational work with absolutely bare minimum interactivity, basically screensavers. I had sort of naively skipped the “But is it a game?” discourse and was like, “Oh, I’ll just stick it on itch.io. It can be a game!” Is it a screensaver? Is it a game? I don’t know. I don’t care! I don’t have the background where I can make a statement bookended somewhat nicely by other developers who were reacting to that space of what is a minimum, viable game and what defines interactivity. This was an era of drifting and wandering games, and I get the sense that those two fronts kind of looked similar from the outside. I was grateful that there was a conversation there. I think that’s why Pietro reached out about that work, but I don’t really know. [laughs] I was honestly touched. I was like, “Yeah, absolutely! I’d love to be involved.”
Can you talk about the development of the game? Were you given any particular direction by the exhibit, or had you already made the game at that point?
No, I made it from scratch for them. There were some limitations. I think it had to be made in Unity. That was the first time I used Unity, so that was not a small thing. I had to learn to use the program to produce this game. In fact, I was grateful for the mission because it was on my list of things that I really had to pick up as a tool. It had to run on mobile, which is no small thing for a game that depends on hundreds of textures, so that sort of idea of compression and optimization was pretty much on the forefront. Those are all technical limitations. In general, they were like, just send us some proposals: What are you thinking about? What is compelling to you? I sent them a bunch of things and they were like, “Okay, we’re interested in this one and this one and this one. Which one of these might you want to pursue?” And I was like, “Oh yeah, this one.”
“Because I went through an arts school education and had made computational-based work and drawings, I think maybe luckily, I missed the whole ‘What makes a game, a game?’ discourse. I wasn’t around game spaces in 2012 or 2013 discussing playability; I was in art school, thinking about sculpture.”
Beyond the endless greenhouses of The Worm Room, a lot of your works incorporate things like bots, procedural generation, and other automata, such as moth generator, which you did with Loren Schmidt. What interests you about procedural generation and bots?
I can talk about how I talked about them at the time, which is a little different than how I think about them now, although they certainly have led into each other. I think there are a couple of components of procedural generation and bots that really drew me to the space initially, and some of those have stayed really important, while some have been replaced by other things. I was compelled by work that exists durationally inside of a human timeline, so things like Twitter bots and other performative entities that can exist at scale with human beings in online spaces. I was particularly interested by this as a side effect of social media phenomena, where there’s an application programming interface (API). People use the web interface at sort of the same scale as the API, and there’s not necessarily any differentiation. This was all before the 2016 election scandal; it was all a little bit naïve, like, “Isn’t it amazing? We live in the same scale as robots on the internet! Nobody can tell!” There’s that classic New Yorker comic with the dog at the console and the dog is like, “On the Internet, nobody knows you’re a dog.” That’s also true of bots. There’s something about that that I thought was neat, as a computational performance that is lived with. It’s like having an artwork in your house, and it’s like seeing an artwork on your feed periodically, that happens over time, that can shift and change, and that can be interacted with in the small scale as a thing that I’ve written. I thought that was fun.
My opinions on that have sort of changed, partly because the world has just turned, but also because I think I have lived more deeply in what it means to be a person that’s tied to a platform, particularly to a corporate platform that has a lot of control over that artistic space and holds bad political values. This isn’t just Twitter. This is true of every social media megalith, that they have impacted human life in a way that is not so much tied to communication and the internet. There are lots of spaces that do well about open-source things like ActivityPub, Mastodon, and the Fediverse. There are ways to do things without necessarily being evil, but some platforms that have historically hosted a lot of those works have sort of fallen in my heart. So, I’m a little bit more skeptical about that kind of ideal of “Yes, let’s exist at scale with our art and our robots and our social media feed.” I think it’s very important and very hard to keep from doing the work for Twitter where it’s like, “What if I made a beautiful thing on your platform, and then gave worth to your platform, and then did free advertising for you, and… oh no!” The internet is never a neutral space.
“I was making weird, lightly generative computational work with absolutely bare minimum interactivity, basically screensavers.”
But that’s a little different than the core of the thing, which is procedural generation, like changes over time or chance operations. That has this long trajectory and history. Humans have been interested in chance, storytelling, fortunes, and fate. The Dadaists were very much into meaning-making and cutups. In a lot of ways, my interests in those spaces are similar to that type of “But what will happen?” What if I built a system that I can’t fully understand, and I press go? What will come out of the other side? That has remained compelling to me, but as I look at the work that I’ve produced and the work that I’m interested in producing with those same systems, the thing that stands out is less chance but more curation. It’s not random—it’s never random—I have made a very careful system that exports “moth” at the end. It’s constrained. Building systems that are constrained in interesting ways, finding boundaries and edges that make the whole potential of the middle compelling and different, is more an act of curation than of random chance. It’s about understanding what those limits and boundaries are, and of pulling in material from the wider world that can interact with the work you’re producing in compelling ways.
I’m glad that you brought up some of those ideas because I watched this really great, wholesome talk you gave in October for the Roguelike Celebration where I think you touch on some of the things you just said. I was especially struck when you described your work as aiming to make “an intimate internet.” What do you mean by that?
[laughs] So much. We talked about the early internet a little bit, like the homepage of your Neopets coded in HTML. There is a potential to fall into a nostalgia trap when talking about that stuff; it’s all rose-tinted. For me, I was a kid, and I think it’s rose-tinted for a lot of people, especially given that the internet we’ve got now is so profoundly not that same space. It’s a space that we can barely access. Unless you’re a very competent developer, we generally exist on the surface layer of the internet. If you can excuse my potential movement into nostalgia, there was something that was different with the early internet, despite all of its problems of accessibility and cost. There are many, many problems in the early 1990s and 2000s with the internet too, but it was homemade almost entirely. It was made by individuals—maybe an individual working for a company in their little department for building the web—but HTML and CSS were written by hand. A lot of the content for those websites were from people who got something up on their home computer running as a server, or got a GeoCities page, or joined Neopets and built something for it. There were so many weird little corners where you could construct things of your own. That type of interaction with that space, that sense of ownership and construction, is something that I feel—even as I would say, a competent web developer—is distant from me now. The internet feels like a space that I go on, interact with, post an image to, or type a thing, but I’m not constructing and actively engaged in making it. And I think that’s a loss for that space.
“It’s like seeing an artwork on your feed periodically, that happens over time, that can shift and change, and that can be interacted with in the small scale as a thing that I’ve written.”
I use this metaphor in my classes about the “bad landlord problem” of the internet, which I think is maybe a good summation of where we’ve ended up. We’re all tenants with bad landlords in websites that we can’t really control. It works on a lot of levels in that we have limited options for customizability. There are things in our lease that say, “You can change your header and your icon, but you can’t redesign the site. Don’t put up wallpaper. Don’t paint. All you can do is hang a picture on a nail, and you better patch that hole when you leave.” And so, we have limited customizability in our aesthetic vision of the internet, which is easy to write off, but is also real. That’s a poetic interface that we all deal with, and not being able to customize it to our own wants and desires is a tragedy. It’s not the way the internet always was. This is also true in things like content moderation or our ability to block and report people. The toxicity of those spaces and our inability to be safe on large platforms is also a bad landlord problem of not installing a lock on your door, or having all the windows open in the house. There’s no chance to figure out what safety feels like for individuals, but we sort of have two options that are set down in this set of constraints in our lease, and we can’t build our own home. We have no capability of ownership unless you are somebody who has a strong tech background, which is another type of privilege.
All of those are an anti-intimate internet. All of those constraints push us away from feeling personally comfortable or personally close to the structure of the thing that we’re interacting with, and we end up like ghosts in those spaces. I don’t feel like we can really live there right now, which is not necessarily how the internet had to be and might not be. I think there are a lot of interesting projects and works—both artistic and technical—that are pushing for a space that is very much not the one we’re in, so fingers crossed!
I was also thinking of some of the other games that you’ve made. Under the collaborative label Withering Systems, you developed this game with Loren Schmidt called inflorescence.city, released in two volumes. The first volume is a procedurally generated catalog of an imaginary city, complete with news articles, census information, library acquisitions, and so on. The second volume includes an explorable map with cardinal directions and stories you encounter. How did this project come about? What were the ideas driving this?
In the first iteration of that project, we were both producing little code poems, experiments, snippets: these little things that interacted with content in a particular way. They would make a paragraph of text that would generate one thing, and we were trying to figure out a unifying theme to publish them. We were thinking about or looking at zines, poetry anthologies, and that kind of history of chapbook publication. We could each contribute five to ten poems, stick them in a document, make sure they all run together, give them a seed from the get-go, and call it a day! That’s how the first volume of inflorescence.city basically came about. We worked on it for not very long, maybe a month and a half to two months, and it was a container for a lot of these little experiments that we had been producing.
“All of those are an anti-intimate internet. All of those constraints push us away from feeling personally comfortable or personally close to the structure of the thing that we’re interacting with, and we end up like ghosts in those spaces.”
The second volume of inflorescence.city is similar, but with a very different unifying theme, which is that of movement. We thought about the first volume of inflorescence.city basically as the generative textual detritus of a city, sort of in the format of a newsletter or a newspaper, but also kind of just in the format of a website. The second one is a lot more game-like in a lot of ways, although both of them are interactive at different points. The main conceit of the second inflorescence is that you’re navigating through this world geographically, and in each position you’re at, it uses a deterministic seed to generate the world. Even though it puts you at a different point every time it loads, it’s always the same world. It’s very big. It’s infinite, theoretically—beyond the human scale of being able to move around and recognize this or that room or whatever.
The seed of the position you are at is always the same, so if you put in that seed in the browser, you will end up in the same place geographically. There will be the same little code snippets around you. I can’t remember how many there are, but there are little code poems that all use that same seed as their generation key. So, if you go to a point on the map, just north of you there might be a moth, to your left there might be a stair, and down bottom, there might be a little description of a window. They are all using the seeds of those locations to create the content. What this means, and part of the reason that we did this, is that we could go and find an interesting space and then share that little piece of text, and then if someone were to paste that in, they could go to that spot. Even though it’s not an MMO, it’s still a shared experience of a world. We were interested in how to produce shared, stable spaces online in a low-tech way, not in a “We’re all in a room together!” but rather, “Oh, have you tried checking out this spot called ‘dog?’” If you type in “dog,” it will translate that into the little code and send you to that spot. You and other people can go there, and you can all see the same thing. That was very compelling to us.
“We thought about the first volume of inflorescence.city basically as the generative textual detritus of a city.”
Another game that you made with Loren Schmidt is Untitled (the flower game), which the site describes as “half drawing machine and half bullet-hell.” It’s a really beautiful game, like a kaleidoscopic version of the classic game Snake. What were the ideas behind this game?
This is more Loren’s design trajectory than mine. It’s not a big part of my trajectory or my heart, but Loren plays a lot of bullet hell games, and so we were talking about that genre and the potential elegance of bullet hell. There’s that bullet hell game, Ikaruga, where you can change the polarity of your ship. It’s a classic. We talked about pattern matching and survival in patterns, and the visual opulence and splendor of bullet hell games, which is just so extreme. We talked about the abstraction of living in patterns as opposed to dealing with the individual enemy, and I think I was working a lot on drawing tools and drawing games. I probably enter all of my various mediums through a drawing lens, as that is my most foundational practice. It’s the thing I return to over and over again, so I was thinking a lot about drawing computationally, drawing on the computer, and how to make interesting marks. Drawing on paper is really just a trace of movement, and tracing movement in a digital space looks quite different but can still make compelling things. We collided those two trajectories of thought in Untitled (the flower game).
It was a funny project. We prototyped it in a couple of days and basically set it down for a year. We would work on it for like an afternoon every couple of months. So, we had this prototype that was basically done, and then it would be like, “Oh, let’s add one little thing. Let’s do sound. Let’s make a web interface, blah blah blah.” It took years to release it, but it was very casual work. In some ways, it’s an atypical project for me in that I feel like a lot of my work is grounded in—I wouldn’t say heavy-thinking—but has a little bit of that type of background. I’m a researcher, and then I’m like, “Oh, maybe I’ll make an artwork about this.” The amount of reading that gets bundled into a lot of my projects, even the ones that are very cursory or feel simple—for better or for worse—is real! That project is not that way. It was very experiential, which is a nice way to work. I guess that’s why you get collaborators; they push you in new directions.
Nature figures prominently in your work in the form of gardens, moths, flowers, butterflies, briars, etc. What is the relationship of nature to computing and technology?
That’s a really interesting question. I feel like there’s a lot of ways that I can go with this. Personally, nature, plants, and the wider world is a large part of what I care for. It’s certainly the landscape that I feel comfortable in. There are parallels between computational work and the ecological world at large. In some ways, that divide is a little artificial in my opinion. Computers are rocks that think. They’re rocks that we have inscribed with enough microcosms and strange little switches, that are run with electricity, and that can do logic, which is fascinating. That sort of separation between a computational object and an actual object is… meh. It’s a semantic line at best. Similarly, code and computational algorithms look a little bit similar to some of the physical constraints of the natural world that we end up with too. It’s not easy to simulate the biological growth model of a plant, but it’s also not impossible. There’s chaos in that world. There’s chaos in computers too, but they also follow patterns. They grow in ways over time that are similar to the ways in which we can model things. Often, when I’m at a loose end with a computational problem, a source of inspiration is biology white papers. How is this solved the old school way? [laughs] “What did this plant do when it was growing? Oh, that’s a neat, weird solution. Maybe I can apply this. Oh yeah, that works!” There are parallels also in a mathematical, and computational, and logical model of the world too.
“There are parallels between computational work and the ecological world at large. In some ways, that divide is a little artificial in my opinion.”
And then there’s the third thing, which is as metaphor. We talked a little bit about an intimate internet. We talked about feelings of ownership and home on the internet, and how distant that feels now. There is a capability to use beauty and the aesthetic strengths of nature to push for that potential space of ownership and home. When I think about a feeling of home, I think about a feeling of care that comes bundled up with caring for my landscape, my plants, my community, the animals around us, and an ecological concern with the Earth in general. That metaphor can and should bleed into digital spaces because they’re a part of that community too. They are the primary ways that we interface with each other. They are the primary communication interface, at least, in a lot of the world. And they are spaces that are often sort of stark, flat, and distanced from the messy, dirty outside in a way that maybe damages the way we think about them and their potential, and our potential to care on and through them.
This maybe gets at this relationship between nature and computing, but I discovered that you made a collection of all the rock models in Skyrim. What did you learn from that endeavor?
Big shout outs to Pol Clarissou, who actually ripped those from the game for me because I don’t have Skyrim. In truth, that was a collaborative project because I was like, “Please, could you do this?” Pol was modding the game; if you haven’t seen that project, it’s highly amazing. It’s called Dreams of Hurling Your Entire Face Into a Rose Coloured Plastic Fern, which is a really beautiful mod. I had intended to collect rock models from a lot of games. I stopped at Skyrim partly because they did it for me. I was like, “Oh yeah, here are some good rocks.” That came from a trajectory of 3D model collections in general, not from game spaces so much but from online repositories of creative commons, 3D models, and things like Thingiverse. I was making these books, and each book compiled one type of model. So, there’s a book of hands, a book of beds, and a book of houses. In 2017 or 2018, whenever I was looking, I would find every creative commons 3D model named “hands” on the internet that I could print in a book. This is actually a small number. You’d think there must be an infinity of hands in the creative commons on the internet, but in fact, there’s really only about 100 to 400 of any given thing that I could find that was licensed for me to use. Doing these collections, I was interested in what a hand was for 100 to 400 people. What was this thing called “hand” and not “left hand pointing?” What did those 3D models look like and contain? The folks who uploaded them, I also gave credit in the back of the book.
But that project with Skyrim sort of came from that question of like, “Oh, what would this look like from a direction that wasn’t creative commons, and that wasn’t 100 to 400 different people? What is this from a design space?” The thing about rocks, minus their textures and materials, is that they’re really just artistic blobs. I think they’re very compelling. There’s this desire to make a stacked blob that looks natural enough to be like an edifice, or a rock face, or a boulder in a field, but not too designed. They’re these weird 3D objects, and I find them quite compelling.
“When I think about a feeling of home, I think about a feeling of care that comes bundled up with caring for my landscape, my plants, my community, the animals around us, and an ecological concern with the Earth in general. That metaphor can and should bleed into digital spaces because they’re a part of that community too.”
Related to that, you have another project called Mushy, which you describe on its itch.io page as “a free asset pack of neural network generated isometric tiles.” What was the motivation behind this project?
I talked a little about curation and dabbling with procedural work, and I feel similarly about neural networks in that I find them profoundly disinteresting unless they are curated. In my opinion, they tend to make an awful lot of mushy noise, right? But mushy noise isn’t necessarily a weakness. You can create a landscape of mushy noise that you can then continue to build from in ways that are compelling and interesting. This project maybe came from understanding the constraints of neural networks at the time. They’ve made leaps and bounds as I was working on it. Even last year, they continue to grow. And they really are pattern-seeking; they’re good at picking up patterns. Things like isometric tiles—which have a sheer bottom that is always set in the same angle—have constructions on top of that kind of rectangular space, and are a really good match for the type of patterns they learned. I then went back through after training on large amounts of data and selected the ones I thought were good. This is a lot of them; there are 824 in total. There’s 440 in the original pack, and then I stuck up the rest. It’s not a harsh curatorial set, but I probably made like 20,000, and pulled from those to be like, “Oh, this one does something odd. Ah, that one has this weird echo.” The other thing is that they came from this creative commons space, so they were almost entirely trained on creative commons tiles for other people to use in their projects. It going back to live in that type of space felt important as this kind of reciprocity of care and attention.
What did you hope that people would do with the Mushy tile set? Have you seen any interesting projects come out of it?
Yeah! There have been a couple. I am always on the lookout for more. Paolo Pedercini released a game called Lichenia, which is great. It’s like a perfect match for the feelings that the tile set contains. He uses the same phrase that I was going to say, the Anthropocene. That phrase is divisive, but he calls it, “A city building game for the Anthropocene,” which is very much contained in the ethos of the asset pack and the interests of my work in general. There have been a couple other projects that I’ve seen people working on. I can’t think of one that’s been released off the top of my head, but every once in a while, someone will tag me in something on Twitter and be like, “I’m working on this game with this asset pack,” and I’ll be like, “Yeah!” [laughs] That’s really how it should go and live. They don’t need to tell me they’re making a thing. Better that it go be in the world.
Earlier, you had mentioned something quite interesting, which is that you thought of some of your games as screensavers, and I noticed that you had a couple of projects like screensaver collection, as well as an individual screensaver release called 761 stairs and 1300 plants, and then your other game mirror lake, which you describe as “also kind of a screensaver.” Why do screensavers interest you as an artistic medium?
Well, we talked a little about minimum viable interaction. Because of the tools that I use, which are computational in nature, I get bundled into the “interactive artist club,” and I do make interactive work, but not really. I mostly make work that sort of plays around itself: the bots, the procedural stuff, books, screensavers, and even some things that I think folks very generously call games that feel moderately interactive. Mostly, it’s sort of happening at you or to itself, and you can witness it or not. I think that’s something interesting to me about screensavers, that they are these weird artifacts from a different technological era, where they had this problem of screen burn. And what do you do to fix screen burn? You can run ambient programs, and then it’s like, “Oh, if an ambient program is running, might as well make it cool.” In some ways, they’re like this amazing standard of artwork on screens in people’s homes, well before digital art was considered a real field in which you could buy work. There were certainly people working digitally, but it’s still a fight to sell a digital artwork to a gallery. It’s not an easy thing to do, but people certainly do it. And yet, all through that era of CRT screens, anyone who had a computer and a CRT device probably ran these in their homes, ambiently in the background, performing artwork, which is amazing! [laughs] Like, in what world??
“All through that era of CRT screens, anyone who had a computer and a CRT device probably ran these in their homes, ambiently in the background, performing artwork, which is amazing! [laughs] Like, in what world??”
I find that space very compelling, like, “Okay, there’s a sleeping screen in our home, and it has this weird tech limitation, so we’re going to put this artwork on it, and it’s going to exist in space, and maybe we’ll watch it sometimes. Or maybe not. Maybe it’s just playing for an empty room.” And I think there’s a parallel in that kind of poeticism with a lot of the work I make in general, which may or may not need to be witnessed all the time. Actively working in that space and producing screensavers from 2018 to 2019—this era of flat screens where they’re not a requirement but where they can still decorate the screens we have in our homes—feels fun and useful. I think a lot of my work in general is aesthetic in its foundation and is interested in beauty, decoration, and the baroque possibilities of technology, which is not maybe a normal thing in a design conversation. However, I see people moving that way and using our devices to the best of their potential, including as artworks with which we live in our houses, and that’s important to me.
I’m thinking now of your talk that you gave for the roguelike celebration, where you urge the audience that when working with datasets and other corpora, what’s most important for you is to “work with sources you care about.” What are the sources that you care about?
Gosh, it’s a wide world. [laughs] There are so many things, but I think the directive of that question was really urging people to rethink the massive training dataset as the standard. It’s very common right now, partly because of the tools that we have and what is needed to train them, just to let them loose in the wide world. It works to a certain extent. The elephant in the room of that talk—I think I brought it up once or twice as being discussed in computational text circles—is GPT-2, which is this incredibly competent neural network training by OpenAI. You might have seen it; it very recently was retrained to be the AI Dungeon 2 that has been making the rounds in social media. That project is a really good example of what it does well. It was trained mostly on Reddit and with some articles, and you can retrain it and tune it to a particular media set, like text-based adventure games, and it adapts to the semantics of a new space. In some ways, that project is more towards my directive of “use the thing you care for,” which in that instance is text-based games. With this type of thing, the more care and consideration we can put into those training models and in the corpora that underpins our projects in general, the better.
All this to say, my own selections of what I care for will change from project to project. I will often approach something like, “I know the feeling of what I want. I have an idea of what this could be, and I have to find material somewhere that is free for me to use. First, that material should not hurt anyone if I pull or use their information, and it should be accessible. Second, it should hit those aesthetic qualities on its own.” That’s really important. The work that you produce out of whatever computational device you create is probably not going to be that much better than your source material in the end, no matter how many constraints you put. Finding something that is clear, lovely, uses interesting language, and contains information about the topic you want to discuss, all of that gets bundled up into that thing.
“I think a lot of my work in general is aesthetic in its foundation and is interested in beauty, decoration, and the baroque possibilities of technology.”
For me, I often look to the spaces that I find compelling in general, which is old internet forums, dusty personal websites that haven’t been updated since 2002, photo archives up on Photobucket, YouTube videos that were never intended to go viral, white papers, and highly specific Wikipedia articles that were clearly written by one person about a topic for which they care. Those types of specific spaces have bundled in them something of a human, not the mass of humanity that gets anonymized in this kind of “We trained a thing on all of this stuff.” You can see the briefest touch of one human being in the mass, and I think those are the spaces that I find most compelling, both online and offline. You can see those things in the world too, like notes on lawns that clearly fell out of somebody’s backpack, the grocery shopping list. There’s the community bulletin board, or the radio in the middle of the night when the DJ runs out of material and just decides to chat for a minute or picks up a phone call. Those intimate moments of broadcast, of finding a person in the world, not a machine aggregating all of people and turning it into a mush that can be used to sell stuff. That is the very exciting potential of that space, and it’s what I return to when I need both inspiration and solace.
One thought on “Player Two: An Interview with Everest Pipkin”