On a technical front, one thing I’ve struggled with often is on OSX the opengl support stopped at a certain point, so certain types of coding like compute shaders are impossible. For me this was one of those “aha” moments that LLM tools provide , that I could just ask, “help me design a system to run compute shaders in Metal inside of an openframeworks application.” I’m sure I could learn the ins and outs of metal shaders, and do this on my own, but within 5–10 mins I had a working framework with interoperability between metal and OpenGL, and it made sketching and ideating so much faster. As a creator I’m trying to find ways that AI can be a tool or an unlock to new mode of practice.

To me the most interesting thing about Physarum is how playful and dynamic the compositions are. Sometimes I think about how much time I could devote to a given algorithm or function. For example, I sometimes think I pretty much have dedicated my life to sin(). I wake up every day thinking about how to use sin() and cos() to make art. Not all functions are like that, and not all algorithms — like Delauny or Voronoi. In the case of Physarum I had the feeling like you could easily dedicate your life to see what stories it could tell.

Fluid

One again, I fell back on the fluid train, a topic I revisit over and over again (it seems like every year now). I love to make images that feel like a water simulation, like milk in coffee or oil on water, and they are created just using displacement after displacement in shader code. I also tried to push them in weirder and weirder directions.

Glitch work

Continuing on the theme of revisiting work, I was showing another artist how to work with glitch and I found myself revisiting a form that has brought me a lot of joy. These glitch images are a result of recursive subdivision and often create pretty fragmentary and textile type images.

One thing I love is just going back to an old idea and seeing it again with new eyes. I’m not always sure I’m making anything new, but I’m new, my eyes are new and I see everything differently.

I also revisited cone gradients for a large circular print that was on exhibition over the summer.

Dots

I love working with dot patterns, halftones and other pattern type approaches. I find the textures it produces fascinating, and in these sketches, I approximated a circle through a series of points of color that essentially paint a circular shape. Each of them has their own orientation and pattern, and the layering produces some interesting effects. I also experimented with non linear distributions, having really large and small dots, warping the dots, and seeing the various ways I could bend or break this approach.

One example, that was aided by this post by XOR about randomness, was trying really trying to push the scale and warping of the dots. I love how this feels like a ball of organic shapes. We’ve move out of the halftone space into something that feels really alive and frenetic.

Lenses

I also revisited lenses, producing images for print:

Circles and Rectangles

I always think of this quote from Vera Molnar, an artist I admire and teach about, “I have no regrets. My life is squares, triangles, lines.” She spent her life creating computational artworks rooted in geometry that use the tension between order and disorder that are full of warmth and humanity.

I think about this quote constantly when I make work using simple geometric shapes. In 2025 I spent time looking at circles and rectangles, trying to find a softness and an airiness in them. For me these pieces are really about light. With the rectangle works, I was thinking of things like Tadao Ando’s Church of Light. And with the circle works, I kept coming back to Turrell pieces.

I made a ton of sketches involving boxes and circles

Line blob / New Yorker

One shape I’ve always tried to simulate and strugle with is a kind of form you see in tape delay machines. I love how these blobby shapes are rigid but also they fold in on themselve and touch to actually negotiate space. I’ve tried to simulate this using my traditional blob techniques (particles that repel away from each other and are connected by springs) and I find they self intersect and I can’t get them to actually touch cleanly like this.

To solve this, I turned this time to box2d, which has an interesting feature called joints, you can connect objects together and constrain their angles:

After some experimenting I got a look I was really happy with:

I wound up using this technique for an illustation / animation for the New Yorker for an article about AI. I made this blobby line compress into the shape of a brain (I actually had to make something a mold for it so it would really get brain shaped!).

Light

Finally, I closed the year looking at light again. When I started sketching in 2016, 10 years ago, I began with light. It felt kind of nice to end thinking of light, revisiting some of my favorite recent techniques and see how I might see them with new eyes.

One obsession I come back to again and again is water and in particular finding ways to explore light on water or in water such as caustics. My studio is near the water in Brooklyn and I like to take walks and look at the east river up close. Whenever I see light on water, I always try to take a photo with two ideas in mind (a) the knowledge that I can never truly capture this complex moment that I’m witnessing and (b) the hope that I can capture the feeling in the picture.

I think when I’m sketching to make something that resembles water, I’m thinking a lot about feeling and in what ways code can convey those feelings. Here’s some water inspired sketches I made in 2024:

I was really happy with how painterly these came out:

And this almost feels like a photograph:

These are just a tiny selection. When I get obsessed with an idea, when I really feel like I’m jamming, I often produce dozens and dozens of outputs. I’ve said it previously but I feel like my job is almost a nature photographer and I’m trying to create a giant contact sheet. I feel like we often need multiples to appreciate the gems and the moments that resonate.

Img2Img

Like many of us (I assume), I have mixed feelings about AI. Certainly tools like LLMs have made my job as a coder *infinitely* easier. It’s astonishing to take old code, say in one language and ask for it to be ported to another, or ask for help implementing a new feature or debugging a problem. But I also feel slightly ambivalent tools that make hard things easy, like writing or drawing or coding. Some parts of life are supposed to be hard — they are the things that give friction that helps creativity blossom. What I try to counsel my students on is that they should try to find ways to use these tools to automate the boring parts of their work but protect the interesting and meaninful parts.

I’m also slightly anxious as someone who makes images and videos for a living to see how quickly tools for generating images and video are advancing. It’s a weird world and getting weirder. I do try to experiment and see if there’s anything fun with these tools I could try. One thing I did enjoy in 2024 was stablediffusion img2img pipeline, which I was able to get to run at about 1 fps on my macbook pro. I found it really captivating to put in a prompt, such as “tilt shift photography of manhattan” and watch a webcam image slowly morph into a blurry tiltshift image that changes as my body changes.

A funny thing that kept happening is I would trigger apples video responses which happen when you do certain gestures, and I loved watching how these would get remixed by the img2img prompt. Here I think the prompt was something like “1950s modern abstract painting”

Interference

I love images that have a certian amount of chaotic energy, and one way to get that energy I’ve found is to add patterns across different points of interest and see how those patterns intersect. Moire patterns emerge I and I always try to find ways that I can work with these patterns but in a way where it doesn’t feel mathematical, as some pattern work can feel. My main interest is always the same: how can I make images that have energy, that feel alive, that don’t feel too mathy, that convey feeling and express something about optics or color.

In addition to interference I also did many studies of ripples in 2024. I like when you can do something patterned but have it feel painterly. I was later commissioned by Bvlgari to make a similar style of image for a campaign. Here’s a video they shot in my studio.

I also tried some very organic forms, they seemed almost like hills or landscapes

Lines and Boxes

I did some experimentation with SDF line functions and painting colors using lines. I often return to simple forms like lines, squares, and circles just to see what is possible and as the starting point for compositional research. I do all of this work in shader code, and I find shader programming pretty relaxing. I have setup my shaders to reload when I change them, and I find I can work in an improvisatory way, adjusting and adjusting until I find something new. Code art is the kind of art that really works with needling, tiny changes can have a big imact, and shaders give me plenty of space to needle.

In addition to lines in a grid, I also put the lines in the form of a box, and experimented with the movement of the box. I like it when it feels painterly, like illumination, or when the box feels like the starting point for something bigger.

Lenses

I did more work with lenses, focusing specifically on movement and animation. I love the glassy textures and organic forms these make.

Boxes

Finally, as part of a project I released in 2024, I focused on box intersections (I wrote about the project here). In particular I was intereseted in how to make complex visual forms with simple geometry and the different ways of documenting light phenomenon. I love images where things get super abstracted, and where the light almost breaks the composition.

I’ll end with some favorites:

In particular I started to experming with ways of making the light extend outward, in different angles. I also thought a lot about different visual effects, like how lights look at night, or how to create forms that feel soft or watercolor-esque.

I love it when the results feel inexplicable but also optical — as if we are looking at something real through a strange lens. For example, having chromatic aberation in the flares:

I also did lots of simple experiments, trying to think of different types of layouts and ways of coloring

I found that I really started to fall in love with directional light, where you see the light kind of move. I love the shadows and soft areas you can find.

Also I really love these wild ones — where I kind of leaned in very heavily to the light outward distortion. They feel almost like particle accelerator photographs.

I love it when images have a kind of kinectic energy to them. Sometimes when I’m sketching my favorite discoveries are completely accidental. They happen by just pushing and pushing and pushing until things break, but sometimes they break in beautiful ways:

Finally, this approach gave me the impression or feeling of reflection on metal. Sometimes when I’m in an elevator that has brushed or shiny steel, I get fascinated with how the light bends across a metal surface.

I also tried combining lenses and light. Sometimes my favorite discoveries come from just putting two different things together. I feel like my sketching process is often combinatorial, trying a+b then a+c, etc. I think maybe the job might to create a kind of grammar, and then experiment with all the different forms that grammar can create.

Circles

I always seem to come back to the circular form. For one thing, it’s one of the simplest shader functions to write (2d circle sdf), but also I find it often gives me starting point for graphical exploration. I often place with the edges, or try layering circles with different strengths to see what I get. I did a series of sketches that explore softness, nesting, focus and color. Later several of these were printed as part of a show at Unit Gallery in London. When I work with still images, I’m constantly thinking of how they will print and often times delighted when I can hold an image in my hands.

Ripples

Ripples were also part of that Unit London show, and I found myself ending 2023 sketching with the ripple form and iterating on it. I’m a bit obsessed (and have done multiple projects) trying to create the feeling of how light would move on water. In these images, I tried to explore various ways of representing a ripple and give the

Hills

I spent time in 2023 thinking about hills, and specifically trying to create forms that undulate like a hilly landscape. Often I’ll start with a basic concept, and then try to add things like shading and lighting effects, so the forms feel more dimensional. Even if the images are not about light, I try to figure out how to use and manipulate shading in some way to give the feeling of light, such a light coming through the layers or light bounding off the forms. I love how some of these feel like layers of cut paper of physical objects.

blob shading

In 2023, Ialso spent some time looking at blobs and finding creative ways to color them. I always love making 2d forms feel 3d using shading. Here I tried some subtle and not so subtle techniques.

I also worked on a variety of blobs for a comission from m+ museum called the Atlas of Blobs, in which I asked a series of people to name and respond to my blobs. I also projected those blobs at the museum and on their outdoor screen.

I also spent time in 2024, making more refined blobs.

I really love these extruded blob packs — the dashed lines give a very pearly vibe:

Color Band Studies

In 2023 I continued my color blind research. I produced many sketches related to these striped color forms. In particular I tried to inegrate sunset colors and explore how different forms would work in bands.

Later I was able to take use some ideas in these studies to a work at Outernet in London, with music by Daito Manabe.

Finally, I don’t know how to classify these, but I continued to make many images in 2023. Here are some of my favorites:

I love trying all different combinations out — almost like how a photographer would have a contact sheet to focus on the best examples, I feel like my job is to make and capture, make and capture and then later find the best selections.

I also experimented with a 2d version of the same ribbon:

I have a fascination with iridescent colors. One of my favorite things to photograph when I’m out walking is the kind of oil spots you see on city streets after rain:

I always wonder how you can create them computationally. I don’t have the mental space to write about them here, at some point, I’d like to write about my favorite computational techniques, including color, and specifically the ways in which I explore algorithmic color generation. At any rate, here’s an example of my attempts to channel some of the feeling of those oil slicks in an image

Light Research

I think of 2022 as a year of light research. My partner at the time, Molmol, was developing her neon practice, and I remember every time I’d come to the studio I’d see these lights reflected on the wall and I was inspired to try to find a way to simulate those lights in code.

I found some shader approaches, such which worked well as a starting point. They were slow, but with some tinkering, such as using jump flood I was able to improve speed.

When I work with light, I found that I often gravitated to simple grids, since I saw these lights almost as giant pixels, like the kind of image you see when you zoom into an LCD screen

I was especially interested in some of the visual qualities that emerge from these kinds of images, for example, the kind of shadows and glows that make a 2d iamge feel three dimensional, as if we are looking into a kind of giant maze from above

Coachella

A highlight of 2022 was working on an audio visual installation with Daito Manabe at Coachella. I was able to translate many idea of my sketching into a 30 minute audio visual installation. The video focused on different themes of nature, sunrises, flowering plants, water, and it was really enjoyable to develop this together with Daito, who is an extremely gifted musician and media artist. We’ve worked together for years and I’m always amazed by the way he can fly in from Japan, put on headphones and compose music for several days straight.

One of the scenes I really enjoyed working on involved 3d point clouds of different trees, and it used a shader based depth of field effect to slice the point cloud in different ways. It ws really nice seeing a tree slides vertically, you could see travel through the trunk and then out to the edges and leaves of the tree, seeing a familiar form in a new way

Horizon(te)s

Another collaboration in 2022 was with an artist I really admire, Iskra Velitchkova. I did a full write up here , but as I look back, the sketches I made on that project really resonate with me. I spent a fair amout of time thinking about how to portray a horizon line and create work with feeling and depth.

I started actually working with lines, since Iskra worked with lots and lots of lines. I didn’t go in this direction, but I love these sketches

Here are some other sketches from that time — focusing on the horizon line in different ways

Some of these straight gradient images I feel are really special. I didn’t use these simple ones but I love it when I find a pocket I can return to later

I also loved these very soft lines, which I used in the project:

One interesting look I found on this project involved multiple boxes, which felt almost like portals to me.

I also fell in love with these box-y glitch images, which I found sketching and also used in the project

Inexplicable Images

In 2022 I made countless images, and tried to make things a weird as I could. I find it hard to write about some of the images because I am working almost without thinking. I keep changing and changing things to see what happens.

I really love making fragmentary images — images that chop up colors and create fields of color.

Fragmentary and wavy is a nice combination:

Lenses

In 2022 I started to experiment with adding virtual lenses to my color compositions. These lenses are like magnifying glasses and push the colors in different directions.

Sometimes the lenses can make a very painterly / fluid look

EEAAO Songbook

I was approach in 2022 by the film studio A24 to make the visuals for the songbook (sheet music) reader to the movie Everything Everywhere All at Once. Each song had a corresponding artwork. Since the movie is about transformation, I used the same piece of code, and tried to push it in multiple directions based on the feeling at that moment in the movie.

Blob Physics

I also explored new techniques for simulating blobs, here using the angles of line segments (their dot product) to determine how strong or weak the repulsion forces are. This allows blobs to have longer, thinner “arms” which I found slightly cheerful and very rewarding to play with.

Finally, I generated thousands of weird images in 2022. Here are some favorites:

Also a nice comment someone left in 2022:

This approach is inspired by Juerg Lehni’s project Hektor, a graffiti robot. Because of the the dynamics of the movement of a robotic spray paint can, the paths it takes can only be smooth curves. Here you can see the kind of loops and swirls necessary to draw typgraphy:

This is a technique I come back from time to time — for example, here in a video I made for the band OK GO, together with Tim Gfrerer

Band Studies

Two of my favorite functions in GLSL arefloor and fract, which are essentially tools to take a number, like 1.39 and return the integer part (1) and the fractional part (.39). I use them often to take something I’m applying, like a noise function, and manipulate it slightly over a grid. I found right away, I could make forms that kind of reminded me of the the Venetian blinds I had on my bedroom as a child, and how the car lights coming through them would make interesting patterns.

Later, in 2022 I did a series of these videos as a project for Verse called Color Blind Studies, and we installed them in London. All of those pieces are inspired by this series of sketches. These we also installed in the Toledo Museum of Art last year. I love seeing sketches at different scales.

Additionally, I was able to revisit these themes making artwork for the IBM headquarters in New York a few years later:

Blur

I also experimented with blurring, as a way of created depth and layers with animation. For these videos, I loaded a font, and blurred every character multiple times (created for example, 50 different version of the letter E with different blur levels), so I could cross fade between completely in focus to completely out of focus. I love the depth of field effect, and some of the ways that multiple blurs give different cues about depth and where to put attention.

Light Travelers

I had several of these studies of light traveling that in the end became the basis for an album cover. Sometimes I use daily sketching as a way to iterate and develop an idea, and see how different graphical forms might work. Smaller clients tend to be ok with that, but when I work with larger brands I have to be more caerful and there have been a few instances where I’ve crossed the line and been asked to remove sketches. I think because I post with very low friction (if I make something and find it interesting, I try to share it right away) it can become something of a grey area.

Circle Blob

Usually, when I do blobs, they are made with springs and repulsion forces, but I also enjoy sometimes making blob shapes out of larger circles that are connected and repel away from each other. For a lack of a better term I call this “circle blob”, and it’s a form that always gives me alot of joy to experiment with. To me, they are some what reminiscent of an Alto vase, and I love to explore the different relationships you can trace out by which direction you walk around the circles to make the closed form.

I also experimented with blurring, and created cloud like forms that way. I’m always attracted to the relationship between soft and hard things, like a sharp edge and a blurry middle.

I also did more traditional blob sketches

In particular, I’m drawn to this blob inside of blob inside of blob. Most computer graphics approaches are additive, adding layer on layer — almost pushing out from the screen, and I love graphics that feel like they go inside and instead of outward.

Circle Push

I enjoyed taking the push / pull research I have written about in earlier write ups and exploring how they might look with circles instead of of rectangular cells. One thing I discovered and really enjoyed was how the circles create more organic and three dimensional forms as they exceed the bounds of the cells they are drawn on. I wound up with these hills and organic shapes, which I accentuated with shader code, adding gradient and other lighting effects.

sometimes they felt like multiple moons:

Spot / Color Research

In 2020, I found it was nice to go really deep in certain directions and explore a graphical idea in all the various directions it would take me. I like to start with something simple, like gradient shapes intersecting

Sometime they make really nice animations.

here, I am attracted to cone gradients and how they might move up and down and point towards the center

the points of color can also be lines instead of cones

Sometimes I would find a certain look, almost accidentally and it really strikes me. I felt this look gave a sense of 1960's illustration — they feel almost watercolor and a bit pyschodelic — and this discovery made me really happy.

I also took the same approach and tried to get weirder and weirder

Some of my favorite images are ones where I step back and try to get very simple. I often oscillate between instense noise / chaos and simplicity

Sometimes I find a certain look I like to explore further, like this chromey feel

I really love this palette, so explored other foms this way

Finally, I made hundreds of images trying to see how weird I could get. Here are some of my favorites from 2021

When I’m explaining these algorithms to students, I always point the Bruno Munari book, Drawing a Tree, which discusses how rule systems, like combining the “Y” shape can lead to a branching, growing form.

I really enjoyed taking these rules and pushing them in different directions:

Atmosphere

In 2020 I also discovered a clever algorithm for simulating the colors of sunset and atmosphere in glsl. This code is actually quite cool / interesting to experiment with, because the constants involve things like the size of the earth, the position of the sun, and other figures relating to the atmosphere. I’ve found playing with these leads to quite interesting results.

I found these sunset colors striking and tried to figure out different ways to push them.

Whenever I work on algorithmic sunset colors, I’m remined of a classic project I love, called Sunset Solitaire, by Joe McKay. In a live experience, the artist attempts to match the changing colors of the sunset with a digital approximation. I love the tension between finding things in the real world and their mathmatical / computational equivalents, which often pale in comparison. Even with the differences (and discrepencies), digital faxcimilies can help us attune our senses and see things in the world in a new way.

Later on, on a project called 100 Sunsets, I found this color research really useful. This project explores how sunsets might look on water.

It’s often not clear when I’m sketching what a given topic will be useful for, but I like the idea that idea or techniques will echo and rhyme in interesting ways.

Circle Curves

Often when I’m sketching I start with a simple gesture, like a line that can only move in a circular path. It starts a circle and at some point, it transitions into another circle piece, and it uses these combined pieces to move in a certain path. I then expand on this gesture through a series of graphical and geometric transforations. Here’s some variations I did in 2020:

the last one became the basis for something you might have seen somewhere sometime from some company.

Objects and Cutouts

One thing I had alot of fun with in 2020 is the use of physical elements. In the studio, there was a pile of tiny clay spheres and I photographed and used them as elements to animate. In addition to the clay spheres, I also used illustrations and photographs of rocks and ink. In most cases,these images were the basis for a kind of particle system, and experimented with trails and other layering approaches.

When I work with organic forms, I’m reminded of an early flash work from Yugo Nakamura where he used clay forms in animation. As a young creative coder, I was really inspired to see natural forms used in a computational way and always return to these types of experiments which informed me on what code could be:

Poisson Light

One day, when poking around github and looking at different graphical libraries, I stumbled on a library called Rendu and was struck by an algorithm for blending color called poisson filling which extends a color outward, blending it with other colors.

I used code created by Lingdong Huang the brings this approach to openframeworks, and experimented with these kinds of soft colors

Body Augmentation

Working with Molmol Kuo, I augmented a set of dance videos, using the body as a starting point for a new type of costume or augmentation. We were trying to see how computation could extend the human body and make more expressive forms.

Blobs with many points

In 2020, I became quite fascinated with different optimization techqniues that would allow me to create blob shapes with thousands of points. Whenever I do particle particle interactions, I wind up making lots of debug screens that I love.

In particular I was trying to solve problems of how to have blob shapes split apart and merge back together, and more optimized particle / particle interaction. I wound up making blobby letter forms

also because I had optimized the particle system, I made videos where I interacted with it using my hands.

I also did fun experiments with blobs (or blobby lines) and light.

Type / Body Gradients

I’m a big fan of distance fields, and work with distance images and SDF functions in shaders constantly. In this case, I wanted to try arbitrary shapes (not just regular graphical forms like lines and circles) so I computed the distance images / SDF images and pass that through to the shader. In this case, I explored a variety of type and body forms.

That’s a wrap on 2020. It’s been fun to revisit and share the process.

Sketches like this are great strtings points for what if questions. What if there were 10 balls?

What if I normalize two balls instead of one? I tried adding a third ball and even a fourth ball until things broke.

As an artist I love seeing when things break. I feel like there’s a kind of “volume knob 11” aspect to daily sketching, and I always like the feeling of starting simply and then loosing control.

Pulled Blob

One of the forms I was really happy with in 2019 is the pulled blob, a blob whose trail or extension shrinks to a point. This is just an extension of an extruded blob, but I spent time thinking about the taper functions and how to create an elegant cone shape. I was inspired by forms like a trumpet, and later I learned about mathematical forms like Gabriel’s Horn, a shape with finite volume and infinite surface area that feels somehow relevant to these sketches.

This pulled blob later became the basis for the cover art on Max Cooper’s Unspoken Worlds album. I was really thankful for how it turned out and to see these shapes in a new context. I can’t remember our conversations but my vague impression is that Max saw these shapes as listening devices, like two ears, and I love that — almost a reverse trumpet.

Blob intersections

I’m quite interested in the various ways you can made blob shapes, and one technique I experimented with in 2019 was looking at the intersections of a long line or set of lines and the interior shapes they make when they cross. I then took those inner forms and rounded them off making blobby shapes.

I tried for example, looking at the intersection of shapes, and seeing the new shapes they make rotating in 3d:

I also experimented with rotating letter forms:

IG Filters

I’m not sure when it started, but in 2019, I found my way into making experiments with IG filters, using spark ar. These were centered around the human face, and explored playway ways of extening or augmenting the face. The nice and suprising property of instagram filters is that the moment you published them, flocks of people would try them and post them, and you could get immediate feedback. Later Facebook / Meta sunset this feature, but while it lasted it was a source of creative freedom and play I deeply enjoyed. I had so much fun watching how people used these.

Unrelated to ig filters, but on the wavelength of cutting up the body, I made some fun sketches in 2019 involving the human hand, and drawing a trail of it over time:

Body Abstraction

One topic that never seems to bore me when sketching is using the human form as input. I have a tiny test clip of a human body I always fall back to. I think it was made by Memo Akten decades ago to test computer vision algorithms, and I often use this as my testing ground for anything involving the human body. In addition I like to use my hand as input. Here I was interested in exploring how I could abstract the shape of the body to make it out of a simple set of curves or angles, applying a sort of graphical constraint. I often like to push things to a far place, like lower and lower resolution, or here, larger and large constraints to see what essential truth remains

I was parciularly font of this “low resolution” body. I love finding things just on the edge of recognizable.

Color Ribbons

I typically create blobs as particle systems, where the particles are connected by spring forces, with additional forces that push the particles away from each other. The springs are connected in a loop, so the form is closed, and depending on various other forces it can expand and contract and fold in on itself, but often it avoids self interaction. If you use the same physics, but instead of making a loop, make an endless line, you can get an elegant curve, that I then extruded into a kind of ribbon form. In particular I like to find ways to use color and shading to make interesting forms.

Color spots

A common technique I return to over and over again, and revisited in 2019, is making images with color spots. This is essentially shader code, where you define the color of a pixel in relationship to a series of spots of color, that can move around, and be shaped an manipulated in different ways. I find always take this sort of basic starting point and remix it, adding layers on top and otherwise seeing how far I can push it. These sketches are some of the more simple version of this approach.

I also made work focused around a center point or lines of points

AR Body

In 2019, apple released an addition feature in arKit which added body segmentation mask, similar to the kind of mask you might see in a program like zoom for blurring the background. This meant it was possible to get 2d information about the body from the mask, and then integrate that data into 3d. I did many experiments with bodies leaving trails in 3d:

On a similar front, I experimented with Kinect Azure, released in 2019, and doing similar layered body things using a Kinect

Body Color Segmentation

I use openframeworks primarily when I sketch, and I sometimes use opencv / computer vision techniques in different ways to create compositions. In this case, I combined opencv with a color segmentation technique to draw video in a fragmentary way.

One thing I really loved was adding different textures to the shapes and seeing how recognizable a face might be.

Color Painting

One technique I enjoyed experimenting with involves “displacement” shaders, where you draw an input image into an output image in a shader, and you add minor displacement, such as a noise function. This essentially move the color pixels by a certain amount. Technically, this is done with shifting or altering the texture lookup. If you do this repeatedly, and take the output of one frame as the input to the next frame, you can actually make colors move and spread. The results are very organic and I’ve found really an enjoyable place for experimentation and play.

It even works with webcam footage:

I later was able to translate these ideas to an interactive piece for WeTransfer, called color push which allows people to paint and push paint using these same techniques.

Finally, it’s a small thing, but I loved these blob lines I made in 2019. I’m always a fan of images that mix curved and angular things, and also give a different feeling of depth. There’s some really refined about these images and I love when I stumble on a technqiue or appraoch that looks really different.

The story of how I started printing is actually quite funny. I was nominated for a Cooper Hewitt Design Award in 2016 (or near that time), and at that time, you needed to supply them with an actual physical book of your portfolio. I had no idea how to make a book! Scambling at the last minute for the application, I bought a fancy printer from Epson and hand printed a book for them. I didn’t win the award and the printer sat unused in my studio. Every day I would come in and it would look at me like “you lost!” Until one day I tried printing still images of my work, and I was really blown away.

I had studied printmaking as an undergraduate and I love the feeling of doing all this work and then holding a print in your hand. Printing on the Epson brought those feelings back to me — I felt like I was back in the printmaking studio.

I was also reminded of how printing things helps you see things in new ways. When you print out writing, you can copy edit it with a pen and mark it and often see mistakes you didn’t see on a computer screen. It’s the same way about printing images, you see them in a different way, off the computer screen, holding them in your hands and hanging them on the wall to live with them.

One form of sketching I did with prints in 2018 is around light and reflection. These images were an extension of earlier work related to light but not rendered at much higher resolution.

I was asked many times about my printer and paper setup, so I’ll just mention them here: Epson p800 and Hot Press Bright. While I created a store and sold prints online, many people also asked about how I printed (which I plan to write up about soon). I kept thinking, “I should be selling printers not prints!”

For some reason, many of my printer tweets would go viral (and I loved the responses I’d get). I have this theory that we’re so used to seeing digital images on screen in such vast quantities that it’s hard for them to have impact, but the moment you see an image in space, on the edge of a printer, in someone’s hands, on the wall, it feels significantly more real.

Weird Type

In 2018 I also released Weird Type, an AR application where you can write typography in 3d. The project was created together with Molmol Kuo, along with Richard The and supported by Design Indaba.

I actually had a really nice section of my daily sketching, where I was reposting other

I’ve compiled my favorites:

Push Pull

A common form I like to coming back to again and again is a grid where the cells in the grid readjust themselves based on some energy. I think the first I saw designs based on this is Lava’s identify for the Beijing Design week:

One thing I always like to do is start with a simple graphical from and push it again and again, like turning the volume knob up to 11 and see what happens. I try all kind of different graphical looks and see the various ways I can break things and try to see what essence or truth remains.

I tried typgraphy

I also fell in love with this sort of blurry effect. I especially love how blurred trails can make 2d work feel three dimensional and how moving objects leaving a trail can tell a bit of a story

Metaballs

I did some futher work with Metaballs. In general, I’m ambivalent about using “common” computer graphics techniques that can be read very easily, such as voronoi triangulation — metaballs fits in that category. I sort of think it’s my job to push it out of something that’s purely recognizable.

One fun fact is that iOS was constantly autocorrecting metaballs as meatballs whenever I would tweet about this work so I think of this as my meatball phase.

One cool thing I discovered was creating a kind of sketchy forms that are moving from one shape to another. I was interested in how lines could connect the metaball forms. I’m always happy with computer graphics that don’t feel exactly like computer graphics but are organic, sketchy and a bit wild.

Blob / Blob Inners

Blobs for me are a safe space. I’ve been exploring blobs for years in my sketches and I think I will always come back to blobs again and again. In 2018 I did a variety of extruded blob sketches, where I look at how making the blob form longer or extended over time.

I also got very excited about the idea of inner blobs, blobs that are actually openings, and thinking about what colors and shades you can see through

I also explore blob “packs” — how mutiple blobs might bounce off each other

Noise Paintings

I did several experiments that are based on noise. I would use 2d noise (simplex noise), and slice it in different layers in a kind of 3d, creating a dimensional form. I was really interested in how these could move and evolve, and how color and pattern might move across layers.

Particle system (body)

I also explored how particles could interact with the human form. I am often interested in various ways to extend or augment the body. Whenever I have code that works for a moving body, I like to connect it to the my hand with a webcam and see what it looks like.

This internal reflection of particles in a polygon code was also applied to blobs, making illuminated, glowing forms:

In 2018 I also worked illustrations for the New York Times, documenting what your body goes through during different phases of opioid addiction. I was able to use some of this research on that work.

Texture Projection

A cool technique that I enjoyed experimenting with in 2018 is texture projection, where you project a 2d texture on 3d geometry and then rotate it so the texture goes from being legible to illegible and back. It’s the movement from 2d to 3d back to 2d I find quite fascinating.

Every Font

Finally, I did some work with typeography. As a “designer” without formal graphic design trainin“g, I always get a little nervous about picking what font to use. I thought, “why not use every font on my hard drive?” These images and animations are looking at letters or words in every single font I could find. One thing I was fascinated with is all the tiny choices the type designers make — like the decision to make a typeface skinnier or wider, rounding it off or straightening it more. All these tiny choices reflect aesthetic intent and they remind of the sorts of choices I make when coding. I enjoyed trying to sort the fonts (such as thickest to thinnest) and see what happens when I overlayed each font on top of each other.

All in all I think 2018 was a fun year with sketching with a fairly diverse set of outputs and many things I come back to again and again.

Exactly 10 years ago, I posted a Happy New Years sketch, something I created with C++ code and recorded using screen capturing software, on Instagram.

While it was the start of a new year in 2016, it was also the end of a very tough year for me. In spring 2015, my father passed away. I remember that I was not on Facebook or Instagram at the time. I was traveling back from France when I heard the news, and I created new accounts on these services, ones I had avoided until then, just so I could stay connected with friends and family.

I was thinking a lot about his life and creative practice, and my own life (I had recently become a stepfather), and reflecting on a kind of unhappiness I was have with work — a sort of mini midlife crisis. At that time I was working a bit like a musician: working 6–9 months in secret on a project the way a musician might work on an album, and then dropping it at a festival or event. I remember feeling intense disappointment when something wouldn’t do well — like not getting press or follow up invitations for exhibitions. I knew intuitively that I wanted to work more open, stress-free way. When I was a student I just made things. I remember being jealous of the freedom my students at the School for Poetic Computation had, and wondered if there was a way to work like that.

After the new year’s animation I started posting every single day after that and never looked back.

Often the sketches were iterative in nature: I’d take something from the day before, and it became a game of changing it to make something new. I started to think of my days now not as AM and PM but more like BP (before posting) and AP (after positing). It became a rhythm and something I started to look forward to every day.

Fast forward to now, I’ve been doing it daily for 10 years, so I thought I’d try to document the 10 most important things I’ve learned. I apologize if these are repetitive to things I’ve said in talks or in write-ups, but I think they are important, so I’ll repeat them.

(1) Always be iterating 🔃

I think the most important thing I’ve found in developing a daily (or regular) practice is taking the burden away from making something new or novel every day. The blank page can sometimes kill you — starting from nothing is great sometimes but usually, if you don’t know where to start, you don’t start.

My solution is to focus on iteration. My motto is simple: Make something new out of something old. Every day I’m often taking code from the day before, and remixing it to see what it tells me.

When you do something repeatedly you have to take shortcuts. When I think about shortcuts, I always think about desire paths, the sort of paths that can be formed by people walking the shorter route. I think those shortcuts become your style and who you are as a creator.

Whenever I talk about shortcuts, I always show this image where a child had to write I will make better choices over and over again — and they optimize it so that all the l’s are one line. It’s the repetition that leads to shortcuts.

This spring I was in Seoul meeting with a friend of mine, Seo Hyojung, who is an extremely talented creative coder, who also does daily sketches. At lunch she told me something quite touching — she actually started daily practice after something I said to her in a comment (which is essentially this point).

Since that interaction she’s been daily posting for years, and I am amazed at how her work and process have evolved.

(2) Document everything 📷

Since I’m working every day, I’m generating thousands of images and videos. One thing I’ve found incredibly helpful is having a single folder called “everyday” where I put everything I make creatively. These files are like breadcrumbs: they document a creative evolution and are often useful for helping me find my way back to a certain place.

Another tool that has helped me immensely was created by one of my MIT Media Lab students Lingdong Haung called SrcSnap.

It essentially ties the act of screenshotting to an automatic git commit. I found that, so often, I would change code, see something I liked, take a screenshot, and then continue changing the code, overwriting that previous moment. This app sits in the background, watches for screenshots, and when sees a new screenshot stores the internal state of a project in its own git repo tied to that image. If I ever need to find the code at a specific point, I can do so through the images.

(3) Reduce all friction to publish 👏

I remember one of my former students, Evan Roth, made an amazing project as part of Rhizome’s 7x7 project, where artists were paired with technologists to make a unique artwork in 24 hours. Evan was paired with Wordpress founder Matt Mullenweg and they built a Wordpress plugin that plays an applause gif every time you publish a new post (video). It’s such a simple idea, but I’ve always felt like I want to internalize that in some way — publishing should be celebrated, and if there’s a way to make it fun, we should.

In my case, I made a concious decision to not overthink publishing. If I like something enough to take a screenshot, then it’s likely someone else might like it too. Over time, I’ve started to have a better sense of when I want to share something, and I always strive to make the process as seamless as possible. Typically it’s as simple as taking a screenshot or recording, AirDropping to my phone and publishing it. In the past when I’ve had more complicated steps — like using ffmpeg to adjust the bitrate of a video — I found it created too much friction.

(4) You are doing it for you 👉 👈

I really like these lyrics from the rapper JME: “Music is reason I make music. I am the person I make it for”

I think this motto is ideal, and something I always want to have in the back of my head. I’m sketching because I love sketching. I’m doing it for myself. I’m not doing it for attention, likes, or engagement. I’m not doing it to advance my career. Obviously those are worthwhile things, but I think the main motivation should be that I love this medium, that I want to see what I can do and how far my imagination and intuition can take me.

The other thing about this quote is that it reminds me of what daily sketching is for me: it’s essentially a diary. I make things and share them, but at the end of the day these are diary entries. When I look at sketches now I can often remember where I was when I made them, what I was feeling and even what mood I was in.

Something that’s interesting is when people like something I don’t like, or I like things people don’t like.

To me those are the most interesting moments because they teach me about how my ideas can be in and out of harmony with the world. To me those misalignments show me what ideas I want to fight for and explore further. If you only do things that you like and other people like it can be much harder to learn.

(5) Seeing with other peoples eyes 👀

This is sort of at odds with point (4), but I think one super power I always wish artists could have is to see the world through someone else’s eyes. You can only see the work through your own eyes, and it’s harder to know how other people experience it. For me the most revealing and insightful moments have come from people telling me what they see in the work.

I love it when someone mentions an artist I am reminding them of — to me it feels like a door opening to whole other world. I love it when people tell me what they see, what the work reminds them of or even how it makes them feel.

Sometimes I even reteweet things people have said with the tag #artisticgoal so that I can find them again. These kind of interactions are really meaningful to me and have made sketching a valuable exercise.

Sometimes the comments are so funny:

But my favorites are simple like this :

(6) You can repeat yourself ♺

This is sort of similar to 1, but I think it’s also important to stress how useful it is to do something again and again. I think artists should feel comfortable going deep on a topic, and revisit ityear after year. There are topics I return to over and over again — like blobs or ripples — because there is always something new for me to find there.

I’ve been looking at 10 years of work this week and one thing that’s struck me is how valuable it’s been to have consistent themes to return to again and again. Here’s how I was describing it:

(7) It sometimes needs to be quick / It can suck ⏲️

I remember I was interviwing Beeple, another everyday artist once and he said, “It has to be something you can do in 5 minutes, because there will be days, like the day my wife gave birth, where all you have is 5 minutes.” That sentiment stuck with me. There are times where it feels like a fun game to try to do something as quick as possible (but feel meaningfully different).

One of the nice things about doing something daily is that the rythm gives you permission to make things that are not always great. I like the idea that there’s always tomorrow. Giving yourself permission to post things that suck, or that you’re not entirely sure about, is really ok.

I remember when I taught at Parsons School of Design, around the time that the iPhone had just come out, and I’d see students work for months / years on a game for the platform. I always felt that the longer they worked on it, the better it had to do, which adds so much pressure. Part of making work quickly is actually trying to remove that pressure.

(8) Boredom is important 💭

Above all else, it’s important to allow yourself the freedom to be bored. We live in an era where it’s almost impossible to be bored. Waiting in line? you can pull out your phone and scroll. It’s an era of endless distraction.

As a creator I’ve found some of the best moments are the moments where I feel truly bored with my own work — and the rythm of making the same style of things daily can elicit a kind of boredom. I look at the code, whatever I’m making, and it can feel like drudgery to pull something beautiful out of it. In these moments — when I am truly aimless and bored, and sort of hating what I’m doing — I often find something unexpected, something really profound or different. When I look for it, I can’t find it. It’s almost like you have to get to the point where you are not looking for it. Those are the points where you find it.

(9) Silence all the voices 🤫

I’m pretty famous with my followers for not taking any advice about what to do. When people make suggestions, I often respond, “DJ does not take requests!”. I am thankful that people care enough to make suggestions or say anything, really, but it’s very important to me to try to silence all the voices except for one: the voice that belongs to me.

As a creator, there are always voices in the back of your head telling you, “It’s not good enough” or “people will think it’s silly” or “you’re misjudging this.” I feel my job as a creator is to silence these voices (as much as possible) to the point where my head is almost empty.

If you are reading this article for advice, I can’t stress this enough, that single voice needs to be your guiding light. That voice might be your intuition, your curiosity, your love for the medium or the spirit that got you started. But whatever that is, that is something you need to protect dearly, like an exposed candle flame in the wind.

I love writing “DJ does not take requests!” If people are rude I enjoy blocking them. In general, I feel like my job is to try to make sure my mind is truly empty. I love not having any idea where I’m going or what I want to do — it’s at these moments you can listen completely to your intuition and let it and luck, take the wheel.

(10) The Ultimate Goal is Freedom ⛓️‍💥

If there’s anything I want to stress about making something everyday, it’s that this is ultimately an excercise in freedom. When sketching, I have no boss. I have no timeline. I have no milestones. No one’s paycheck or career is in the balance. There’s nothing really at stake besides my time and my hard drive space.

There are those practical freedoms in sketching, but also a true creative freedom. Excited about typography? Want to focus on color? See an exhibit that inspired you and want to respond to it? You can.

To me those freedoms are ultimately what a creative practice needs: The freedom for something to suck. The freedom for it to be stupid. The freedom to repeat yourself. The freedom to get lost. The freedom to truly listen to yourself. The freedom to go deep or broad on something. The freedom to celebrate the medium for what it is. The freedom to make for the sake of making. The freedom to capture your feelings or whims. All of these freedoms are what make a daily practice rewarding. The only constraint is time and the dedication to keep making and sharing.

Sometimes when I would be working on a crazy hard project — pulling an all nighter or just in crunch mode — I’d take a pen and draw a the letters K.G. on my hand, like a tattoo. It stands for “Keep Going”. I think that’s the simplest way to describe a daily practice. I posted something once at the start of 2016 and I kept going.

I guess if there’s one thing I’d say, after all this, for anyone interested in or curious about this practice it’s simply this: K.G.

Box light studies is a long form project I’m launching on artblocks in an auction that concludes today. It represents a view into my daily sketching (for the last 8 years) and my studies of light, geometry, color and texture. Everyday I sketch and I think of the question, “what is the best combination of light, geometry, color and texture that makes me feel something, so that I can share it with the hopes that others feel something.”

First, I should say I have a love / hate relationship with long from work. I find it a hard medium, since it involves random outputs, and so much of my practice involves curation. I’ve said before that in my practice I feel like a wildlife photographer, making images and videos and trying to find the right moment to capture. I think the selection is part of the art form, and with long form you give that up in some way to randomness. It’s a definite challenge but something I think ultimately helps expose algorithms and is a bit of a crucible for generative artists to learn about themselves and their work.

I also have had a personal struggle with artblocks. In 2021, during the height of the speculative mania around NFTs I had a curated dropped planned with them in the fall. I found myself going crazy, if I’m honest, given how much time I spent looking at the random outputs of the system I had built. I love sketching, and I love trying all kind of ideas, and the matrix of possible outcomes kind of broke my brain. I really didn’t like who I had become. I found I was obsessed, and I just couldn’t get my head around what I was making and what it was making me feel. I cancelled my project, which was in some ways potentially walking away from life changing income, because I felt like my mental health was more important.

This is not the last time I’ve cancelled artblocks. In the summer of 2023, I was also working on a curated project (based on this code) and I had the same feeling, I couldn’t wrap my head around what I was making and I just need to the time and space to feel like I was ready. In both cases, I’ve been very thankful to the artblocks team for being increadibly accommodating and supportive of my decisions.

Now I’m finally here (!), with box light studies. I LOVE light. I think light is one of the fundamental things I want to spend my life studying and making work around. I am fascinated with optics. Sometimes I get the feedback from people about my work that says, “your work reminds me of an ocular migraine I’ve had.”

This is because I’m concerned with and make work around the field of optics. How do we see? How does what we see make us feel? In my case, I want to explore illumination and how light moves in mysterious ways. For example, when I walk my dog at night, and I see a car driving towards us, the headlights have these weird fringing. You can see almost a kind of glow. It’s impossible to photograph, it’s impossible to record it, but it’s something apparent on your eyes. I love those qualities.

Often when I’m in a new city, I spend my time thinking about light. For example, here are some images I took walking around in Taipei:

I love getting inspired by things in the world, then trying to code them, and in turn having the way I see the world change.

When working on this project, I chose the box because it’s a fundamental form. I was inspired by artists like Manfred Mohr and Josef Albers.

I was thinking of these boxes almost like prisms, and I love to see how they intersect and relate to each other. A box could be a container, a present, a gift. A box can also define a territory, or a boundary. The nice thing I found about boxes is that they intersect in different angles, and create tensions between 2d and 3d in ways that I like as an artist. I always want to make 2d work that feels like 3d and 3d that feels like 2d. I find it invites your eyes to do more work, and there’s something about the eye->mind->eye journey that I’m interested in.

For the light code, I used a technique that involves a lot of computational work per pixel. Esssentially it’s a kind of 2d raymarching, where the pixel shoots off in random directions to figure out its color. It means the image develops slowly, over time. There are examples of this kind of approach here and here. To help optimize things I use a technique called jump flood, which can be used to turn images into a distance map, useful for speeding up some of the light calculations. Lingdong Huang, who is a PhD student in my research group at MIT, made a great example for this here (which proved super helpful for this project). Other bits of code that helped were the fbo class from Dave Pagurek and Matt Desalaurius’s tiny artblocks.

One thing I was really interested in are all the various ways light can move and bend and translate. I experimented with dashed lines and weird glitches. I wanted to see light bend, and how it could be not just lines that are emitting light but also the relationships between lines and the space around them. Here’s some examples of these qualities I aimed for:

For color I opted to use a color palette that is based on prisms. I tried a variety of colors, and in the end I settled on a palette that seems to be expressive and true to my interest in optics. I didn’t opt for a variety of palettes but rather to see how I could push one set of colors in different directions.

I opted for an arrangement of 3 boxes. It’s a bit personal but this year has been a year of transitions and this summer I spent a lot of time thinking about my family and the changing relationships between my partner, my stepdaughter and myself. To me these boxes were people, who are in and out of harmony with each other. Which are touching each other, but also not fully contained. Some of the boxes are 2d, some are 3d, and some are somewhat between. It was really important for me to explore the tension inherent in a group of three.

Three is very important number to me because it reminds me of my father who was very superstitious and would always make us say good luck in groups of three. He was a professional storyteller, and when he would go travel he’d be leaving as ask for three good lucks and I’d say them and add fourth and he’d run back to the house asking me to add two more so that it add up to a multiple of three.

I love numbers. I think there’s incredibly beauty in math. In this work and in my work generally, I am trying to photograph math. I think of box light studies as a series of photographs. Sometimes I feel the photographic quality quite strongly, with tiny overlaps of light at different angles:

I use the word study in the project title because to me everything feels like a form of research, that it’s about studying the physical world and the world of computation. The work is an exploration, it’s constantly evolving…. I feel like I could spend my life trying to understand and study light, and I’m happy to share where I am now.

I’m so thankful to finally do artblocks and share this work with the world, and I want to send a huge thank the artblocks team (especially Sarah Rossien and Holladay Saltz) for all their help, support, and patience. I’d also like to thank Molmol Kuo, Lingdong Huang, Dave Pagurek, Chih-Yu Chen, Iskra Velitchkova, Lars Wander, Benjamin Simon and Matt DesLauriers.