The Color of Code | Jeff Davis

Few individuals have played a greater role in the recent resurgence of generative art than Jeff Davis. A former painter and professor who has published textbooks on color theory, he is now one of the world’s leading generative artists as well as Chief Creative Officer at Art Blocks. However, unlike many of the current golden generation, he trained as a fine artist at the School of the Art Institute of Chicago. It was in the 1990s that he began using digital tools to develop compositions prior to executing them in analog form. Since encountering Processing, he has increasingly conceived of his art in digital terms and following the rise of NFTs has been exhibiting work in its natively digital state. Ahead of the launch of his latest project, LED, as part of Bright Moments Tokyo, he spoke with Alex Estorick about the art of coding color.

‍Alex Estorick: What is the Jeff Davis origin story?

‍Jeff Davis: I’ve been producing art since the early 1990s, after studying fine art as an undergraduate. I’m trained in traditional printmaking techniques and have an MFA in painting and drawing. At the beginning, I was making very hard-edged geometric work that probably looked digital. In grad school, I got hold of design software for the first time — early digital tools like Adobe Photoshop and Illustrator, which I used for planning and sketching. The digital component helped me to map out the proportions, scale, and color relationships in my work.

Back in the ’90s, I would develop works digitally, but there wasn’t really a mechanism for them to exist as digital works in their own right, so I would execute them as paintings, silk screens, and collages. It took me a while to accept that digital sketching wasn’t simply a preliminary step.

At the time, printing technology wasn’t fantastic and I wasn’t keen on the quality of the digital prints I was seeing. Something shifted in the late ’90s, when I discovered printing technology that was suitable for a fine art context. Around 1999, I transitioned from preparing works digitally to becoming a digital printmaker. But it was still a non-generative and non-computational form of abstraction based primarily on architecture — the skyline of Chicago, for instance. I would take the basic mathematics of my surroundings, simplify them, find the order, and then rebuild them into highly austere geometric abstractions. That led me to develop a geometric framework and a compositional and color framework with which to approach art. 

I then began introducing randomness into that system in order to build things with their own volition. Initially, I would create Excel spreadsheets to make random compositional decisions for me, but that became tedious. Around 2010, I discovered Processing and found that I could build a system to make random decisions that could also do all the drawing for me. I’ve been working that way ever since.

‍AE: I’m trying to understand the difference between developing an idea to be painted and developing an algorithm to be generated. Do you find yourself more influenced by painters or computational artists?

‍JD: Even when I was a painter working with traditional media, I was still thinking structurally and systematically. Over time, I developed an unconscious system with an inherent rule set — right angles, whole number ratios, thick black lines, flat colors, and color relationships born of numeric sequences. I started to think that there were three or four related things that are serial, that always use the same rule set or concepts. It was a natural transition for me to go from my original way of working to thinking about the system set itself. Now I let randomization happen inside of that system, which removes one part of the decision-making process.

I had a traditional art education, which is where the normal canon of modernism and contemporary art influences came from. I was drawn to Pop Art early on in my career because of its bold, flat colors, and thick black lines. That plasticity really appealed to me. I’m also influenced by Minimalism and the idea that you can strip something down to its sheer essence. Sol LeWitt was a massive influence on me, but not so much his Wall Drawings as the Incomplete Open Cubes (1974/1982), which is part of LACMA’s exhibition, “Coded.” LeWitt demonstrated visually every possible outcome within a system while also explaining the entire system by showing all its possibilities at once. That really resonated with me.

However, it was thanks to Josef Albers that I experienced my color theory revelation — that an artwork can be self-referential in what it’s trying to explain. His work teaches you color theory just by looking at it. 

My exposure to historic computer-based, code-based, and algorithmic art is actually very recent. Before NFTs, I worked in a bubble. 

I became aware of Casey Reas after discovering Processing, but at the time I was much more engaged in the mainstream contemporary art world. I was showing my work in more traditional settings in the Phoenix area. I have never thought of myself as a strictly digital artist, only that digital tools were the best thing for me to express what I was trying to accomplish in my art. I was much more interested in the conceptual push of my work and digital was just a tool for me to get there.

‍AE: Your work seems to unite two histories that have tended to exist separately: that of minimalism and conceptualism on the one hand, and that of computational artists such as Vera Molnar, Frieder Nake, and Manfred Mohr. Are you conscious of that hybridity? 

‍JD: Prior to 2019, before I transitioned into NFTs, I thought of myself as an artist and digital printmaker, someone who made physical objects to be put into galleries or homes. My practice was entirely two-dimensional, static, and flat. Even if I was working algorithmically, the end point was paper in a frame and on a wall. What interested me about NFTs was that you could have a digital work of art that didn’t have a physical manifestation. But then the first two years of my NFT career were essentially under lockdown at a time when there weren’t exhibition opportunities. Back then, it made natural sense to create digital work to be consumed online, so I was doing work on SuperRare, Rarible, and, in late 2020, on Art Blocks. 

Then Bright Moments appeared in mid-2021. Seth [Goldstein] reached out and said, “I’m opening this NFT gallery in LA, I think you’d be interested.” People were finally getting vaccinated and getting together again in public. My first project for Bright Moments was Portal (2021), which was my first attempt at animation or time-based work, and the first time I imagined a digital representation of my work in a physical space on a screen rather than as a print. A lot of the work that I’ve been doing with Bright Moments has been more experiential than a singular two-dimensional object.

I’ve drawn inspiration from artists like James Turrell who specialize in light and space, thinking of the digital display as a light-making device. 

‍AE: What do you think about the emergence of long-form generative art, whereby artists design algorithms that are both varied enough and of a consistent visual standard to work across many hundreds of outputs? It must demand different skills to produce vast long-form systems than to curate smaller editions.

‍JD: Before working with NFTs, I was still conceiving of algorithms as a means of creating physical works of art. Because of that I was always limited by how many pieces could fit in the gallery space, and how much it would cost to frame works for the show. Although I was working generatively, I had in mind a series of between eight and twelve works that would work for a solo exhibition. With NFTs, those restraints have been lifted. 

Now that we’re working in a purely digital space, there is an opportunity to go much bigger and think about algorithms that can handle a larger number of outputs without repetition or oddities. The twist is that if you’re going to allow the generation to happen on demand at the point of purchase, it requires a change of thinking. Because of the nature of my work and its simplification, I’m drawn to the idea of having a very tight, finite system and exploring all of its outputs.

Smaller editions are probably more natural to me than massive systems that hold great complexity and pockets where emergence and grails can occur. 

My interest in simplicity and core structure means that I often move in the opposite direction toward a more flattened space. When you work generatively the system creates an algorithmic space into which you inject randomness. Every time you do that, it creates a new possibility inside that space. Flatness consists in a smaller algorithmic space with fewer opportunities and possibilities. Because there are fewer decisions being made, there’s a greater similarity between the outputs inside the space. 

‍AE: Are there any generative techniques that you are particularly inclined towards?

‍JD: I don’t use any mathematical packages such as flow fields or Perlin noise. Because I’ve been painting in an abstract geometric way, my general practice has been to choose programming functions that do the thing that I was doing before. Of course, that allows greater control. 

But really, under the hood, my work has a lot to do with whole number proportion, counting, and arithmetic sequence, both in terms of composition and color. 

The question is how to turn that math into p5.js functions, and then introduce randomness to it. But my randomness is not noise, it’s integers. For the most part, my work expresses the relationship of a whole bunch of integers to one another. It’s a low-level way of doing algorithmic work because it’s very finite with a limited number of things that happen — everything is either true, false or one to 12, basically. That’s how I approach the construction of compositions and the relation of colors to each other. 

I generally like the project to dictate its size. My largest project is Color Study (2021) which has 2,000 outputs, but, since then, I’ve mostly been doing smaller projects. I’m interested in super narrow algorithmic spaces where you can only have a handful of things and before they start getting repetitive.

‍AE: Does color have different qualities in the context of digital art than, for example, traditional painting?

‍JD: I think so. Until the last couple of years, I also taught integrated design and color theory in higher education. I’ve written textbooks: Foundations of Design (2016), which is about composition, and Foundations of Color (2015), which is about color theory. I was taught the original red, yellow, and blue color theory as well as how to mix paints. 

The more I’ve explored color theory, the more I’ve realized that it is a broken color system that actually doesn’t reflect how things work.

The RGB color model is appropriate to light and can be extended to the physical light on a surface, which is reflected instead of projected. Color is also, in and of itself, infinite. Maybe it’s not perceptually infinite, but there’s infinite room inside the world of color. I find it interesting to chop it up and put it back together in different pieces. 

If you’re tightly controlling a system, you’re suppressing emergence. But color interaction also involves emergence. If you are refining a system down to a finite set of colors, the second you put two of them side by side, they create unexpected effects. While color study structures the choice of colors, the interaction that occurs between their progressions and the way they touch each other is unexpected.

‍AE: Generative artists are some of the principal beneficiaries of the NFT as a technology because, despite its long history, generative art hasn’t had the potential for a mass market until now. As a leading generative artist based outside of New York, who nonetheless trained at the School of the Art Institute of Chicago, do you feel like an outsider, an insider, or an oscillator?

‍JD: For my entire art career, I’ve felt like an outsider. I was doing minimal geometric work when postmodernism was exciting, and I was doing digital work when painting was having a resurgence. I’ve always felt like I didn’t belong in the place where I was working. It wasn’t until the emergence of NFTs and art on the blockchain that I encountered other people like me who were interested in collecting this work and understanding it. With the success of Art Blocks, we’re now opening up doors and having conversations with the broader contemporary art world. 

But I also feel that we are engaged in a new way of doing business. It’s not taking what we’re doing and saying: “for it to be real, it needs to fit into the contemporary art market.” We are charting a new course and creating a new model for the creation, distribution, and collection of artwork. 

‍AE: You are a practicing generative artist, but the language you use — that of “serialism” and “flatness” — feels like it’s addressing the very same problems as Donald Judd and Frank Stella in the mid-1960s. 

‍JD: I’m certainly exploring similar concepts, but I’m doing so from a generative, code-based undertaking, so it’s different. My work started with tinkering with Excel and using data tables to draw things. Realizing the inefficiency of that approach, I learned programming in order to automate it. That was a major step in my career because the ability to iterate at speed changes things. I’m still engaged in some of the same concepts of structure and form, but the layering of the generative component is more contemporary and allows for more rapid investigation, and deeper exploration, of those concepts. Maybe on a surface, plastic level, you may see visual similarities. But I’m now able to explore a much fuller representation of these systems and ideas.

‍AE: Beyond your artistic career, you are also the Chief Creative Officer of Art Blocks. How do you balance it all? 

‍JD: I’ve never been solely a professional artist. I’ve always had a job and been an artist, and because I was in higher education, I was teaching young artists how to become artists. Before Art Blocks, I also ran a print publisher called Davis Editions, working with contemporary artists to help them realize digital prints of their work. 

As Chief Creative Officer at Art Blocks, my role involves working with generative artists to help them realize their projects. I’ve always gained a lot of satisfaction out of helping other artists bring their work to life. And now, alongside my wonderful team, I’m responsible for the art that comes out of Art Blocks, managing relationships with artists and the projects that are hosted on the platform.