Waiting To Be Signed · interviews on generative art, on-chain
← Index
Interview // FEB 2024

Jan Robert Leegte

Title: Sculpting The Web
Role: Generative artist
Platform: Verse
Duration: 1h 6m
Hosts: Will & Trinity
Listen on Spotify Guest on X Download MP3
#056 · Sculpting The Web
Self-hosted audio // press play
1h 6m
MP3 ↓

Spot an error? Highlight the words or lines you want to fix — an “edit” button appears, and the panel opens with your selection ready to edit.

Trinity: I think maybe one of the differences between net art and AI is that you have all these people, such as yourself, who have been practicing net art for years and years. And then all of a sudden there's this gigantic and very hungry community of people who want to support the movement. Whereas with AI, I think it's a much smaller niche of computational geniuses who may have been dabbling in the past — it's more about the proliferation and accessibility of AI. What was it like for you coming into the world of Web3 and blockchain? Was it like coming home?

Will: All right, hello and welcome everyone to another episode of Waiting to Be Signed, a very special interview episode. We're joined today by Jan Robert Leegte. And of course, Trinity is here for this amazing interview. We're talking to an early net art legend, Jan. Trinity, how's it going?

Jan Robert Leegte: Good. Great to be here. Thanks for having me.

Trinity: Thanks for joining. It's always a little intimidating when we have people who've been practicing art in a very serious way for longer than the last 3 to 6 years on the show.

Will: It makes you a lot harder to research too, because we actually have to get into the archives and see what you've done in the past.

Trinity: We're not calling you old. For the record, we're just calling you experienced. Thanks so much.

Jan Robert Leegte: I just turned 50, so I do feel a little old, but I've got a couple of years to go. I think it's a special age. Will reached out after hearing you on Artnome, and Jason was specifically saying you should engage with artists. You were like, "whoa, we don't know" — but I said, no, it really is a good idea, and you do well.

Will: Yeah, we just get scared from time to time. But here you are, and you're not scary at all. Hopefully we can let everyone learn a little bit more about you. The place we always start is asking you to introduce yourself and tell us about your background in art and coding.

Jan Robert Leegte: I started doing architecture in the early '90s and enjoyed it, but it wasn't for my brain — you need to handle very complex issues as an architect. It's a very complex, layered profession, and I'm just not that type. I had a more philosophical interest in life in general, so it was too complex and too applied for me. But I loved dealing with space, the social aspects, the body, and being able to design that.

So I shifted to a more experimental corner and moved to art school — a logical move. Of course, you end up in the sculpture department, which is sort of mini architecture, and I started making installations and performing in them. They became condensed studies of autonomous architecture in a way. I had a great time, but it felt a bit like repeating history, because a lot of that had been done in the '60s and '70s by these giants I really admired, and I was looking for my own edge.

I'd always been interested in computers. I grew up with a Commodore, and my dad brought home an old Hewlett-Packard — I remember the first game I ever played on its teeny little screen. I was touched by the magic of them, and had my first PC as a student, but never engaged with it profoundly, just playful stuff — Word docs, a bit of gaming.

I was very lucky to go to an art school in Rotterdam — this was '95 — that had heavily invested in computers and had a lot of Silicon Graphics machines. I don't know if anybody remembers what those are, but they were these teeny little supercomputers, probably as fast as a cheap phone now. They were used for 3D rendering in the early days, and nobody used them — there was a whole workshop full of them sitting idle. But there were also classes in interaction design, working with Director and then Flash, and we even had lessons in web — learning how to make web pages.

I did all of that as a sculptor, because I was very interested in learning it. It was through the interaction design teacher — who miraculously knew about net art — that things clicked. This was '96, which is crazy for a teacher to know about that. I've never met a teacher like that since. He showed me Jody's work, jody.org — I don't know if you're familiar, but they were pioneers of digital, web-based, net art, going back to the early '90s. They actually have a big installation in the permanent collection at MoMA. I'm always surprised people don't know them — they're up in MoMA! Very impressive work.

So I got shown that and thought, wow, this is mind-blowing. But I still didn't really get into it until I got my own web hosting package and uploaded my first experiment. That moment — pressing the button, watching the file go to the web server, then visiting it in your browser — became mythical to me. That's when I thought: wait a minute, this is a completely new field. I think it was the public aspect of it — the realness of it being out there, people can see it, it's not on your computer anymore, it's elsewhere, it's out there, it becomes public software in a way. People can click on it, and it really resembled architecture — this is physical stuff you can make. So I instantly saw it as installations: installations of scrollbars and buttons and spaces and frames and color fields. I never really looked back.

That was 25 years ago, but it's such a fascinating field because it moves forward faster than you can keep up with, and it's still a relatively small group of people exploring it. So that's my history.

Trinity: When you talk about using it as a form of installation — have you seen the movie Hackers?

Jan Robert Leegte: I think I have. Is that an old one?

Trinity: The one with Angelina Jolie.

Jan Robert Leegte: Yeah, exactly, the old one.

Trinity: There's this visualization of hacking as almost moving through a cityscape, finding all the different files and servers. It's kind of like an installation in a way — you can see it, you just need to know where you're going, similar to how we might use Google Maps to find a public installation on some corner.

Jan Robert Leegte: Totally. I was a big fan of Snow Crash — similar vibe.

Trinity: The best book, in my opinion.

Jan Robert Leegte: The best book ever. One of my top, top, top ones. That installation feel is very much where my work comes from — a very material feel for the screen and the software specifically. When you bring up Google Maps, I think about what it actually is — a layering of a classic map, an interactive layer, a database — a composite of different media. That's what software is: you can stack stuff and stick it together, add all these different layers, and it becomes a whole new material. It really feels sculptural to me — like really making stuff, although it's on the screen.

Will: You mentioned the installation quality of uploading your own website, and I came across an interview where you spoke about the internet broadly as a public space to perform in. Trinity and I remember — we're not as old as you, but we're pretty old for crypto — the early AOL disc days. We were right on the edge, but not experimenting in the same way you were. As someone who's been active in creating and documenting this space for coming up on 30 years, what have you observed has changed, and how is that reflected in your work?

Jan Robert Leegte: That's such a big one. I see all the big movements reflected across Web 1, 2, and 3. Web 2 was never really my medium, because my work is so material and not social per se — there are wonderful artists who cornered that social aspect of the web, in what's called the post-internet phase, art-historically. So you have net art in the '90s, post-internet, and now blockchain art — three big movements in internet art.

A key point of the second one was bridging into the gallery space — that was very defining of the post-internet movement, and something I was doing since 2000, 2002. My first translational piece put scrollbars into the gallery space as projections on wood, and I've never really changed that. It's this game of physical works across all kinds of media, but always centered around the network computer as a subject — adding to this research, this narrative, trying to dissect what this machine is, artistically and as a phenomenon in life. So I hop quite easily from medium to medium.

That's easier for me coming from art and adding the internet, because working with early internet meant zero audience — just a few of my buddies doing it, and maybe three curators. It was a teeny little world that slowly added a few more people but remained very underground. To have a life, you also had to operate in the art world, and you wanted its recognition, but it wasn't very interested. That was a hard uphill battle — it forced you to translate into the languages of other traditions, like sculpture, installation, and performance, which I loved. I think it's great to make those connections across media.

Then Web3 came, and that was a first real emancipation of net art. Suddenly it had a base, a societal value — though people disagree with that — but there was a massive community supporting it, understanding it, wanting to learn. It became like a circle for me, coming back to making net-based work as a core project, plus all the quirks of the movement — blockchain comes with its own peculiarities and specific technologies you have to engage with, which makes it exciting.

For me, it's amazing to see that history unfold. That's the great thing about working with the internet — it constantly surprises you. Who thought Web3 was coming? Five years before, nobody had a clue, and it blew us all away. AI is next — it's going to change everything, the way Web2 and social media changed everything. Ask people five years before, and nobody really sees it coming. I find that fascinating. This internet has a mind of its own, and we just have to go with it.

Trinity: I think maybe one of the differences between net art and AI is that you have people like yourself who've been practicing net art for years and years, and then suddenly there's this huge, hungry community of people who want to support the movement. Whereas with AI, it's a much smaller niche of computational geniuses who may have been dabbling in the past — it's more about the proliferation and accessibility of AI. What was it like for you coming into the world of Web3 and blockchain? Was it like coming home?

Jan Robert Leegte: I wasn't that early — my first drop was in November 2021, just after that crazy summer, which I missed entirely. Before that I was just trying to survive. I have two young kids, ten jobs, you know the drill — it's hard trying to maintain being an artist full-time, and I didn't have the space to look into this. So I saw it superficially, as many did — just square images — and I didn't see the point. I wasn't anti per se, but I didn't see the innovation in it. That's something hard for the pure Web3 community to understand, but the amount of innovation done in digital art before that is immaculate — in a lot of ways further along than what Web3 was doing at the time. So when I just saw a square image, I thought, this hasn't even started. I was missing all the nice pieces.

It was a tough time — I was heading into a nervous breakdown, had to quit all my jobs, drop everything. In the middle of that panic, Harm van den Dorpel — you probably know him from Mutant Garden Seeder and other drops — came over. He's a good friend of mine, and he'd just dropped Mutant Garden Seeder to huge success. He brought a crate of beer and just kept talking at me, giving me the pep talk: you have to join, this is so fitting to what you're doing. Then he left, and I was thoroughly confused, but I used my last bit of strength to look into it. It actually felt therapeutic to spend an hour or two a day playing in this new field without any obligations.

I did that for a couple of months, gave up many times — tried to learn Solidity, gave up horribly. Tried to connect with Art Blocks, nothing happened. Tried Folia, nothing happened. It can be very daunting when you're fresh in this — how do you even start? You could just mint something on OpenSea, but that won't do anything.

The big thing Harm convinced me of that evening was that the blockchain is just software running on a ledger — decentralized software. You can do whatever you like with it, make generative art, whatever. That flipped a switch for me. I spent those months reading, looking into on-chain stuff — I was immediately fascinated by it. That's the corner I landed in.

Eventually someone from Folia said, "You want to do a drop? I know a programmer who'd like to work with you." I said that would be wonderful. He took me by the hand and explained how the world worked, and we had a great time developing Ornament together. Then one day he said, "It's ready, time to drop." I panicked — whoa, whoa, whoa. But that's when it really started. Suddenly you're in this world you know nothing about, the collection sells out, and you're like, oh my God, what now?

Shortly after, my contact with Art Blocks became fruitful — you reach out, and two or three months later they say, "okay, we want to do something." That's when Window came out. Now it's nearly three years later, and I'm starting to understand how it all works. But it's completely fascinating, and it's swallowed up all my time since.

Window — Jan Robert Leegte

Trinity: The new proverbial upload-to-website, release-this-on-the-blockchain scenario.

Jan Robert Leegte: Deploy contract and there it goes.

Will: I'm curious — so much of your early work, especially in those early web days, was about running your own server and uploading a project as a free, interactive, exploratory thing. Moving into NFTs, from a world where you were basically publishing for free — did you feel any tension around the idea of adding ownership to the internet? That doesn't get talked about much. Obviously it's great that artists can now sell their work, but there's another side to that coin. Some people's vision for crypto is to make everything ownable. Do we actually want that?

Jan Robert Leegte: It's a specificity of the medium — it works better with some works than others. It's not one-size-fits-all, and a lot of net-based or digital art just isn't translatable to NFTs. If you drew it as a Venn diagram, you'd have this big pool of digital art, a smaller island of art within it, and then a smaller subset still that's NFTs, because it's more specific. I can make browser-based art as a collectible series, but if I don't have to constrain it that way, I have much more freedom. And if I don't even need a browser, I can make even crazier things. So it's limited, and I think people forget that in the craziness — like this is everything, this is perfect. It is limited. Pros and cons. It's fantastic to engage with an audience and have that support, but I think the audience around the work is one of the most powerful things here, and it could be utilized more. That's a future direction I'd like to help push — more medium specificity, more experimentation. There's a lot of conservatism too.

The one I found most telling was, of course, the goose from Ringers — one of the most celebrated icons of the NFT generative art movement — depicted as a framed picture of a goose on a wall, which is a centuries-old trope. That's not exactly the progression we're talking about. It shows how conservative the audience generally is, which is normal — it's true in the art world too. Art is supposed to be on the edge, but a lot of it is quite conservative, very formal, not really medium-specific — could've been done other ways, could've been printed. There's so much more possible.

Look at something like Mitchell Chan's Boys of Summer, or Jonas Lund — two wild, coincidentally related projects. Jonas Lund's stacking piece, tied to his home: when he drops something, it changes an NFT. Very playful, very early, and there's so much more possibility there, but everything is still restricted by the blockchain contract format, which is quite limited in code possibilities — not a very dynamic language or platform. So you soon have to mix in other media — server-based components, game engines — and it becomes a hybrid form. I think we'll see more of that, because for purist on-chain art, it's limited. I'll keep making it, because it's great, but it doesn't show the full scope of what digital can do.

Window — Jan Robert Leegte

So that's the double edge of the coin. I'm intensely grateful to have found an audience that actually gets my work — refreshing after 20 years of mostly not meeting them, at least not beyond a small group. The art world has supported me, in my experience, but more out of historical acceptance than real conceptual understanding, because they don't have the reference points in their background. Art world training is very humanities-based; it often lacks the technical or computer-historical knowledge to get those references, to understand the materiality and why doing it this particular way matters. In Web3, everybody gets it. That's been the biggest surprise — I put subtle things into my project descriptions and people instantly pick them out: oh yeah, that's that, that's that. Very cool. And financially, of course, that's great too.

It had more of a HEN feel before — we mentioned HEN a minute ago — more experimental, not aimed at profit, because nobody thought about money; it just wasn't on the horizon. It was purely about the art. It'd be interesting to see if there are corners where that spirit can flourish again — and with the market cooling down, we may see more of that.

Trinity: That's such an interesting topic, and something we've talked about a lot on the show as we've settled into this extended cooled market. How do you think it feels for an artist coming from the '90s, where you'd just release something and it's consumed for free — ownership was loosey-goosey, maybe eventually acquired by some institution, like we mentioned with MoMA — to now having a generation of artists who came in during Gen Art Summer in 2020–21, expecting to sell digital work for hundreds or thousands of dollars a piece? Now we're in a cooled market where things might not sell out, or sell at all, at those prices — you might only be able to release for free or for tens of dollars. Does art need to be financialized? How should artists approach their expectations around that?

Jan Robert Leegte: It's sad, but you probably have to let it go. When we reach equilibrium, I think it'll still be a massive gain compared to before — being able to connect with a new generation on an international market in real time is a huge upgrade from depending on galleries: local, slow, expensive, and limited to a narrow audience of collectors. So it's a big gain, even if it gets decimated. You'll have to find your own audience like always, and work much harder for it. It becomes like the art world, but online — which is probably the normal situation. What we saw was crazy and abnormal.

The speculative aspect will remain, because it was always there in the art world — just amplified, like the internet amplifies everything. So everybody's a bit worried, especially if you're dependent on it. If you stepped in in 2021, you have a skewed view of how this works, because that wave of cropped-up potential exploding was a once-in-a-lifetime event. We might see others, but not that one. Still, it's fantastic to have this market available, and I am finding my audience. Some collectors might stop — I wouldn't blame them, it's an expensive hobby — but there will be interest, new people will come, and they're so much more approachable. I'm quite optimistic about that.

Will: To shift gears a little and ask about some of your specific projects: in your intro, you mentioned AI as something going hand in hand with web3 that you're really excited about, even though we're not sure what its impact will be. I'm not sure if it was your first project using AI, but the one I found recently was Mountains and Dropshadows, which you released on Verse -- a reimagination of a web-based project you'd done years before that served people random images of mountains, probably sourced from a search engine, and imprinted drop shadows on them in a browser. This time you made the mountains using AI instead. What inspired you to revisit that piece, and what brought you to AI? What surprised you about the experience, and what concerned you? We've seen some artists be really vocally skeptical or fearful of what these tools will do to the perception of art -- and not just art, but skill, talent, and hard work. You've clearly embraced it, so I'll let you riff on that.

Window — Jan Robert Leegte

Jan Robert Leegte: I can give it a shot, though I'm definitely not an expert -- there are artists way better at this. But you have to engage with it. It's just part of your culture; there's no way around it, and it's good to keep track.

I first used AI with Web, which was a recent drop, in September, I believe. But the idea predates web3 entirely -- that was my response to AI back in 2020 or so, when the first GPTs came out and I saw the potential for generating code and imagery. I thought: wait a minute, if it can create code, why couldn't the system create its own network of web pages? That was the seed of the idea, and it got shelved for years. Then web3 happened, and Fingerprints approached me about a project -- this was maybe two years ago -- and I said, well, I've got this idea of an AI-generated network of web pages. They said, whoa, that's pretty cool, that'll suit us. So we did it.

It was a purely conceptual idea, which is how I approach AI in general -- very conceptually. I'm not a "tools" person; I don't really use software to make images. I'm making work about the software. It starts with observation, with something conceptual and narrative. So with AI, I thought: it writes code, and HTML is such ubiquitous code out there, it should be really good at it. Let's see what it can do.

I love the idea -- classic science fiction stuff -- of the internet filling its own gaps and starting to populate the web we've sort of left behind. Web 1 is past; we're all living on six URLs nowadays. It's this vast empty space, and the web would start populating itself, which made total sense given that AI is basically one big database of the web -- it's sucked itself up and is now reproducing itself. I also thought it'd be exciting and alienating to stumble into this machine-created web.

So I started working with it, and it really didn't work. I tried and tried. But that was useful -- I learned what AI does. If you're working with text and code specifically, code is so structured and formalized that AI is just not good at creative coding. It doesn't do that. It's an incredibly normalizing tool -- it always thinks in the middle, it can't go to the edges and fringes, because it doesn't have the data or the capacity. It couldn't make freaky websites. Just really dull layouts, every time.

So I started stacking things -- hacking my way around it by layering AI on top of itself, analyzing images, generating text from that, then generating HTML from that. It became a real dinosaur of a process, and it still wasn't that interesting. In the end, I realized I could far more easily just write it generatively in good old human-written code. That coincided with the decision to move it on-chain, which turned out to be way more elegant and exciting -- a whole network of web pages, like a big bang popping out of one contract. I'm very happy with the result, and I learned a lot.

Window — Jan Robert Leegte

Mountains and Dropshadows was an easier move, since I just wanted another iteration of a commentary on online images. The original work queried databases for images tagged "mountain" and stacked them with a drop shadow, mimicking a desktop environment -- commenting on what a landscape image becomes in the context of a desktop: this anonymous, delocalized, beautiful but very sad mountain residing on your desktop, and how it engages with the material on top of it. It became a sculptural statement. Revisiting it, I thought, well, I have to do that with AI now, because the databases are passé -- we're never going back to a stock photo bank when everything's dumped into AI and you can just prompt it on demand. So I did the same thing: prompted "mountains," and there they came, beautiful and wonderful, hardly that different from the originals -- just less quirky, more normalized. You don't get some naked person running through them suddenly, unless you prompt for it. Cleaner, more stylized. But it fit, and became another iteration in this progression of technology. It was a very easy one, honestly -- prompt "mountain," get it back. I spend a lot of time thinking about the concept; the actual execution is very, very small.

Trinity: How many mountains did you pull from all these different databases?

Jan Robert Leegte: For which one, the first or the second?

Trinity: The second.

Jan Robert Leegte: The series ended up being 80.

Trinity: How many mountains went into the model?

Window — Jan Robert Leegte

Jan Robert Leegte: I just used the public model -- I didn't feed it anything myself. The first piece queried Flickr and similar sites, and I didn't want to add any content to the net -- there's enough out there already. I wanted to see what the internet had to say for itself about what an Alpine mountain looks like. And the results are pretty impressive. I actually ended up using Adobe, which was a surprising twist -- I tried DALL-E and Midjourney and wasn't that happy, then thought, hey, I have the full Adobe suite on my computer, they have an AI tool now, let's give it a run. It was really good -- could change the aspect ratio, all kinds of tweaks. This is a plug, I know, but I hadn't expected to be that happy with it. A company like Adobe should be good at this, but they'd seemed behind. Turns out that's where the images came from in the end.

Trinity: I haven't tried Firefly much beyond upscaling once or twice, but--

Jan Robert Leegte: Yeah, stuff like that.

Trinity: The possibilities are definitely endless. Maybe this is a good place to transition into the combination of work and concept -- there's a constant theme through most of your work, I won't say all since I'm not the expert on it, you are, but: traditional web components, past and present. The drop shadow, for example --

Jan Robert Leegte: Mm-hmm.

Trinity: It's a classic CSS class you see everywhere. Maybe a little out of style now, but there was a time in internet history when everything had a drop shadow. And we see the same thing with buttons, JPEGs, windows -- all these pieces that just feel so classically online. Where does that come from for you?

Window — Jan Robert Leegte

Jan Robert Leegte: The way I frame it myself, it's a phenomenological interest in the software. Translated into human terms, it starts simply with doodling -- with a button, dragging a scrollbar up and down when you're sixteen, feeling stuff with your mouse, the edge of a window. A very tactile, material interest in software: resizing it, seeing how it changes, feeling it, playing around with it. That's where the sculptural interest started, I think, and why I chose the interface as my sculptural focal point instead of working in a 3D program making big rendered sculptures. I think that's really where the magic happens.

After twenty-five years I've built up stacks of personalized theory around this, but essentially: the interface is a magical fleece -- it's all we actually have. There's nothing but the interface; behind it is completely inaccessible. I wrote about this for an earlier show -- there was a famous shift in the '60s when programmers coined the term "transparency" for a methodology: transparency meant hiding all the inner workings and only showing the general lines. A paradoxical term -- transparency meaning obfuscation, hiding. That programming concept launched Steve Jobs's mission at Apple and became its pinnacle: make it easy and accessible at the cost of hiding all the inner workings. I found that so fascinating.

Of course we can access code, that's not a problem, but I often go back to Charles Babbage and his difference engine in the 1800s. If you were observant, you could actually watch a computer work -- see it calculate, follow the assembly line. Now nobody can understand how the voltages work in real time. I think that's the big difference. A phenomenology of software isn't about how it conceptually or technically works -- it's about engaging with it in real time.

I've recently started a meditation practice, and I've always been fascinated by John Cage, this Zen artist so much about observation -- what is this, in the now? There's a great difference between writing code and running code: they're two different worlds. You have this text, and as soon as it runs, you have no idea what's happening behind that fleece of the interface, until it crashes and you see text again. It's this magical world of running software, and I'm passionate about what that is. So in my work I'm asking: what is this stuff the software produces? What are these images, like JPEGs -- what are they, what makes them, what are their aesthetics, how does that relate to art history? I keep circling that. It's become my core -- very meta, as you said, Trinity, but that's it. I'm observing. My work is about the network, the computer, the software. That sets me apart a little -- there are other artists who do this, but not that many. Most people see the computer as a tool to make art. I make art about the tool.

Trinity: You're very literal about net art -- it's art about the net that heavily features parts of the net.

Jan Robert Leegte: Yeah.

Window — Jan Robert Leegte

Trinity: It's very enjoyable, but for anybody experiencing net art for the first time, it might just confuse them about the entire genre. Is this work commentary on the internet, or not necessarily? As we move past Web1 and Web2 into Web3, we're really talking about the blockchain as almost an interfaceless entity—it's much more about what's happening in the background. Maybe we can talk about Broken Images. It feels like commentary on Web3 specifically. What can you tell us about that project for listeners who might not be as familiar?

Jan Robert Leegte: Broken Images started when Harm van den Dorpel came to me on a September night in 2021, convincing my little cracked brain to join this party. My first idea was actually Broken Images. I remember telling him I'd love to mint a broken image, and he said, "No, you can't do that. That's not Web3—that's linking it to Web1, that's something different." He was very against that project, which was funny. But honestly, I didn't know what it was yet myself, so I shelved it. It only came into fruition this year, once I understood the technology a lot better and understood what I wanted to say with it.

Then I thought: no, it actually does make sense. One part of the story is commentary on Web3—basically 99% of all NFTs are links to something else, so it's a very vulnerable ecology. If a link breaks, you see a broken image, or a 404. But I've always been interested in the idea that a broken image isn't something that doesn't work—it's actually something else, something very native to the browser. I wanted to make a series using that as material, minted fully on-chain. I love the energy of having something completely on-chain, unbreakable and immutable, but pre-broken—the images are already broken by design. So you get these compositions of broken image placeholders in the browser, stored on-chain and thrown out into the world.

What's interesting is that it works two ways. On one side, it comments on NFTs that can break—except here, they've been pre-broken so they can never break any further. On the other side, the image defaults back to the placeholder, which is something in flux: as you view it on a different browser, or in the future, it will change. It mutates and reflects the technology of its time.

I've done this before in my work. There's a piece called scrollbarcomposition.com, a web-based work from one year after I finished academy—simple HTML and JavaScript, a modernist composition of hyperlink blue, grays, and whites, with scroll bars triggered by JavaScript to move. That piece is from 2000, and it's been educating me for the last 24 years about what this medium really is: a transformative, mutating entity. It's shape-shifted through Aqua Apple style, Windows XP, Linux, phones—constantly changing. Responsiveness is always central to my work: it even changes with the type of screen and resolution you're using. That fluid, morphing quality is such a key property of browser-based and digital art in general—I find it essential.

Window — Jan Robert Leegte

So that piece is changing and will keep changing. I'm curious how it will evolve as UX changes—maybe scrollbars will disappear entirely and it'll go static as a composition. That's fine, and by then everything can be emulated, so you can go back and watch it across all its different phases. I find that absolutely beautiful about working with the computer. Broken Images works the same way—I've already seen it render differently in parallel browsers at a gallery show. People watching on Linux or Safari get a different icon than the one hanging in the browser I like. Web3 collectors instantly dig that. They really love it.

Will: This has come up in a few of these interviews—preservation and longevity with NFTs. Is the image or code stored somewhere on a private server that could get shut off and brick the piece? Could the blockchain itself die? Will web browsers 100 years from now not run JavaScript anymore?

I completely missed Web somehow when it released, but I saw it all over people's best-of-the-year lists for generative art. Looking into it for this interview, it feels like a culmination of everything you've talked about so far. For me personally, it evoked the early days of being online—hunting for hidden links, selecting an entire webpage to see if a single letter or period was secretly a hyperlink to some hidden part of the site. Each piece is a treasure hunt, and I love that you can jump out and see the map of everything you've explored—hugely nostalgic for me. And reading your description, you've been really thoughtful about preservation, making sure it can work moving forward, and that if something happened, people could still rescue and resurrect it with whatever future technology exists.

I'd love to hear your thoughts on preservation in Web3 and net art specifically, and how that intersects with your choices around on-chain versus not, and which blockchains you commit to. I think you've only worked in ETH, if that's right—do you discount other blockchains for one reason or another? How do you think about all this as someone trying to make work last 50, 100 years or more?

Jan Robert Leegte: My bet would be Ethereum—historically, for the team, the decentralization, the market capitalization, all of it. I know people who work at the Foundation, and it's pretty much going to run for a long time. Smaller chains carry more risk of losing support.

I've looked at Tezos many times—I got leegte.tez really early, just never used it actively. A lot of buddies there kept reaching out, "come in, come in," but I didn't have the time. It's done a lot of good projects; I've just been busy elsewhere. Now it seems a bit quiet there, so I don't know if it's the right moment anymore, and a lot of people have moved over to Ether instead. I'm happy with that—I've gotten used to it. I like the bruteness of Solidity; you can define things as deep as you like. That's also why I didn't go into Ordinals—it doesn't have the same technical depth. You could upload a whole script into Bitcoin, sure, but I didn't see the technological or conceptual value in that for myself, so I didn't engage with it.

Window — Jan Robert Leegte

Ethereum is it for now. I have no idea where the rest is going—Solana is growing, but I don't know much about it, and I don't know many people building there. It takes a lot of energy to get into a whole new chain. I did dabble with layer 2 with Coin, which I released this summer—a crazy performative piece, 10,000 coins across four different layers, four different chains. Fun, but I don't know about long-term engagement; it gets messy. You need focus in this world—it's already so chaotic. That's probably why I'm on Ethereum.

Will: Right, other best practices—it seems like you're very committed to on-chain. Anything else you do to maximize the long-term viability of your projects?

Jan Robert Leegte: The safest approach for me, for one-of-ones—I don't have that many, but Broken Images is a good example—is handcrafted individual pieces on my own contract. I use a mutable contract. I do put things on IPFS or on-chain, but I keep them fully mutable, which is against the ethos in a way, but I think it's the better option for the collector. If something breaks, I can fix it—update the code, swap the IPFS repository, whatever's needed.

For truly special one-of-one pieces, especially ones going into museum collections, you'd have a hard time even selling to a museum without that option—they'd ask what they're supposed to do with something unrestorable that they can't access. It doesn't make sense for an institution. So mutable contracts are my best practice. You have to trust me, but serious artists in this space generally aren't going to fool around with that. It's more the temporary, commercial, massive PFP projects where you wouldn't want mutability. But for singular artists, it's the better way to work—it means I can hand over control to an estate, a gallery, or a conservator after I'm gone, and they can maintain the work.

For a lot of locked on-chain projects, they just become part of the fluid language of the net and mutate along with it. I'm curious how a lot of generative art will hold up, because everyone's been making it—from zero skill to very high skill, full 3D gaming shader stuff. Before Web3, nobody would've considered selling something like that to a collector—you'd need a maintenance contract and so on. It's funny how that shifted overnight: unsellable stuff, and suddenly everyone goes nuts for it, even though you don't really know what you're buying. Some of these highly complex pieces could break pretty soon.

The stuff I make is brutally simple by comparison. As I said, I have a piece from 2000 still running beautifully—very elementary. I have a lot of faith in the internet's backward compatibility. Stuff from the '90s still works fine if it's built in HTML and JavaScript.

Window — Jan Robert Leegte

Trinity: Hopefully you didn't release anything in Flash.

Jan Robert Leegte: Don't do Flash. Don't do third-party stuff. We didn't expect in the '90s that JavaScript would still be running this long, but it became standard practice to keep old libraries backward compatible—they don't take up space, so browsers just keep supporting them. It's easier than publicly announcing "we're not supporting that era anymore."

It's pretty amazing how stable it is. I often say web art is turning out to be the most stable medium I've ever encountered—way better than video, painting, or photography, which degrade pretty badly. This just keeps running and running. As long as it isn't Flash—we lost so much with Flash—but GIFs, JPEGs, they'll be around for a long, long time.

Will: We're counting on it. Trinity, one more question, or should we move to rapid fire?

Trinity: I have a related question to what we were just talking about. There's been so much conversation, especially around IPFS versus on-chain versus something like William Mapan's Sketchbook — partially tokenized, partially not. When you're thinking about NFTs, do you see the token as the art, and/or as the certificate of authenticity for the art?

Jan Robert Leegte: I'd say it's the art — for me, at least. That's because I really invested in understanding what the token itself is, and in actually using that. Ornament, my first piece, was already on-chain — it's partially a contract work, partially web, sort of a mutant. Then there's Art Blocks, which is great, but it's out of your hands. Building your own contract is really great fun, because as soon as you're doing that, you can say the token is the artwork — you can consider every aspect of it. That's wonderful, but it's not for everyone. You need to work with a programmer, or be a really good programmer yourself.

Window — Jan Robert Leegte

But it's a thin veil. Broken Image, which we spoke about earlier, is clearly a work made for the token. It exists only there — it's about the token, so it becomes this holistic little entity, a truly medium-specific work. If you're just minting images, though, that's a whole different story — and that's probably what NFTs were invented for. That's the real utility of them. There's this weird subgenre that's created an on-chain, generative, medium-specific practice, but for the vast majority it's just a certificate.

I've thought about adding certificates to my print work, because it was cool and hyped for a while, but I never did it — and I'm probably happy I didn't, because it makes things messy. Then you have two things to take care of, two things that can get lost or broken, and I can't imagine collectors being happy with that. It doesn't seem to have taken off much. I don't think tokenization of art is a big thing, really — I don't see a lot of it. It's still about NFTs, and the digital-native, browser-based stuff is the fun part. That's where we should keep it.

Will: One more rapid-fire question along these lines. One of the themes we identified for 2023 that really came true in this bear market was the hybridization of NFTs with physical pieces — there are platforms, like Tonic, built entirely around that. You've done physical work in the past, taking elements of the net and making them physical. Do you imagine exploring that again — a tokenized piece with a physical component?

Jan Robert Leegte: I'd only consider it if it were a hybrid conceptual necessity. I'm looking into more social interaction within networks of tokens, and I can imagine having components that live outside the token — but it would really have to make sense, and it would have to be transferable. That's the whole point: if you sell your token and there's a physical thing left behind in someone's house, the system breaks. You have to think it through carefully. But there's definitely still space to explore there.

Trinity: One trend we've noticed is that a lot of the best code-based artists of this generation are Dutch.

Jan Robert Leegte: That's so weird. We do have quite a lot of them.

Window — Jan Robert Leegte

Trinity: Agree? Disagree?

Jan Robert Leegte: It's such a freaky question. Why is that? There were a ton of pioneers in Web 1, 2, and 3 — Jodi, who I mentioned, and plenty more. Post-internet, we had a couple of really good people too. A lot of art historians have tried to break their necks over this question.

I think it has to do with — and I hope this doesn't change — we've always lived in a very progressive country. Recently there's been a more conservative shift in government, but traditionally we're very progressive. It's a small delta in Europe, heavily based on trade throughout its history, and that tends to create pragmatic, progressive people who adapt to a lot of flux and import rather than being nationalistic. We're also constantly fighting the water, so there's this dynamic cultural history.

We saw very early adoption of the internet — a lot of influential internet companies, and early punk and hacker scenes in the '80s and '90s that adopted it quickly. I think one of the first internet providers here was started by a hacker group, coming from that alternative scene. And we have a heavily subsidized art culture, which I think is fairly unique in the world — a lot of space for experimentation, supported by the country. That's created focus in weird niches where nobody else would have the support to pursue them. That's changing now, but we've had our share of luck. Beyond that, it's probably just that once you have a few people, they influence each other, and little cultures grow. I'm very happy to be part of that — they're all good friends of mine, and it's been very inspiring to work like that.

Will: One more, and then we'll wrap up. Who would you like to hear us interview in the future?

Jan Robert Leegte: Speaking of generative art in general — my friend who brought me into this, Harm van den Dorpel. You should definitely talk to him. He was doing NFTs seven years ago or something, on Bitcoin — really early, a genuine pioneer. What I love about his work is that instead of making a project in p5.js, he works from Darwinistic, hereditary systems, or Markov chains, or uses social media itself as a randomized generative system. He takes all these different computational systems and works from them. It's super inspiring — he's really adding to the history of generative art, pushing its edges. I don't know if you've talked to Raphael Rosendahl — he'd be great fun too. He has his own podcast and very good opinions on all this.

Window — Jan Robert Leegte

Will: I think Harm has a Verse show coming up.

Jan Robert Leegte: Yeah, he does — right about now, isn't it?

Trinity: January 30th.

Will: The first Verse solo show of the year. All right, add them to the list.

Jan Robert Leegte: You never know.

Will: I think that's a good place to wrap it up. Jan, anything you want to plug — what should we be on the lookout for from you?

Window — Jan Robert Leegte

Jan Robert Leegte: I think I'll surprise people. There are a lot of projects I'm working on, nothing concrete yet. There are some parties I'd like to work with, and we've been in contact — including Verse, we're doing another project with them, probably later this year. So I'll surprise you guys.

Will: Thank you so much — it was awesome to finally sit down, get this recorded, and really interrogate you on net art and web stuff. It's been a pleasure. We hope everyone enjoyed it. That's it for this one. Bye, everybody.

Jan Robert Leegte: Always listen — we're waiting to be signed.

Change log

  • Initial transcript — auto-transcribed (AssemblyAI) and readability-edited.