FOLIO - MATT DESLAURIERS
This was part of an ongoing series of articles that was released digitally throughout July 2022. They were first published in the print edition of the Bright Moments Quarterly that was distributed in July in London.
Malte: Thanks for agreeing to this interview, Matt. I’d like to start by asking you how the series you are exhibiting in London began. What were your first inspirations and how does your creative process unfold?
Matt: I can’t say exactly when the series began, but I think I have worked on it for about a year. One of the things I have been fascinated with for some time is visual poetry, especially concrete poetry. You can’t really read these works like a traditional poem. And I began to ask myself how I could use code in a similar fashion: as an exploration of text, typography and language. In this regard, I am also often inspired by early computer works like Frederick Hammersley’s The Computer Drawings (1969). These graphic compositions were made with ASCII symbols due to the constraints of the system, and the drawings were made with massive IBM 1403 printers.
The London Collection is a continuation of my interests in visual poetry and early computer constraints. Each time the system runs, it produces a new layout of typographic elements to form a woven tapestry of text and typographic detail. The composition emerges from a limited set of constraints: each image is made of only 2 or 3 color layers, and all the graphic detail comes from the arrangement and overlay of ASCII glyphs.
I should mention that as we record the interview, the work isn’t finished yet. And one of the questions I am still thinking about is whether the output should be purely abstract, consisting solely of random letters and glyphs. Or whether I should tweak the algorithm in such a way that, interspersed within the structure, there are semantically meaningful words and phrases.
I started this series with the intention of further exploring my interest in typography: using type as a paint brush, you could say. And the way my creative process develops is that I let myself be guided by the code in many ways. Most of the time, I do not exactly know where I am going. Rather, I let the code go in different directions and this lets me find things that are interesting, things that I want to explore further. It’s almost as if the code shows me things: it is like a strange collaboration with the program. And this leads you to a result that you didn’t expect or intended at the beginning.
That’s fascinating. Poetry has tackled this gap between meaning and the visuality or materiality of language in different ways. And you address this issue on an entirely new level.
Yes, there is something there that I certainly gravitate towards. One of the early outputs of the series got me interested in this question: a distribution of random letters that by chance formed real words and sentences. These surprises continued to happen and that confronted me with the question whether I should explore it deliberately. And I think that is exemplary for how the generative art process works: you get different outputs that surprise you. This allows you to explore different trajectories by tweaking the algorithm in a specific way.
Each work in the series has several layers. Was this something you discovered in this way?
Absolutely. And there are really two parts to the layering.
First, it’s the idea that generative art is more a system than an image. When we see the thumbnails of artworks, we see these isolated images. But what interests me about generative art is that the artist creates a system, a set of rules and parameters. Each time you put an input into the system, you get an art- work. And this artwork can be represented as an image or as a print. But it can also be represented through projection-mapping, VR, AR, textile, machine-controlled printing and cutting, even ceramic and sculpture... So, there are an abundance of ways to represent the output. But the core of the project is the system, and that is something that is rarely recognized.
Without the image, we would have no appreciation of the system. But what the image does is really to reveal the underlying system. The image peaks inside the system, in a way. It feels like you get closer to understanding it, even if you might not, in a technical way. The more images you see, the more you understand the system that generative art is. Each token is a single input in a system, and the system can produce a variety of outputs. The life of the project lives in the system.
Second, I try to emulate real world process- es with code. And the different layers also refer to the process of printmaking. One can think specifically of screen prints and the different layers this procedure requires. My code is simulating this print process and the different layers it requires.
At first glance, the works appear very graphical. Upon closer inspection, how- ever, the viewer realizes that the shapes and forms are constituted by the arrangement of typographical elements. This could be read as reference to code, or as an evocation of ASCII art prior to the invention of graphic processors. But it of course also opens a rich field of associations to the use of typography in 20th Century art. We briefly touched on this point, but would you like to elaborate it further?
In addition to Frederick Hammersley’s The Computer Drawings, I’ve also been inspired by a range of visual poets like Paula Claire, computational and constrained poetry like Nick Montfort’s coded poems and Taper (an online journal for computer poetry), and graphic designers like Ryan Carl who apply minimalism and language in their work.
Moreover, there is also this range of artists that are a constant source of inspiration for me: Anni Albers and her work on textiles. Another thing that inspired me a lot are Dadaist poetry and the Surrealist games that produce chance images or make-believe poems by following specific rules.
In fact, there is an interesting aspect to doing similar things with code. If I would simply create a “poetry bot” and it spits out the outputs on Twitter, the system-based nature
of this would barely come to light. If I lay out the textual elements visually, however, you have this immediate association of the “poem” being system-based. This is really what I am reaching for.
Is this due to the context?
Yes, I think it might be related to that. On Twitter, the context would simply obliterate the system that stands behind the output. But if the output has a visual aspect to it – be it an image, a print, or an animation – it does feel more system-based. And you can really get a feeling that this poem is very computational. The possibility of doing an installation in a physical gallery allows you to create an even better context for highlighting this aspect. For the London show, I am considering using an old computer terminal that would display these outputs.
When we met in London in early May, you shared some of your thoughts around balancing chance and intention in selecting mint 0 for this show. Could you say more about this? How do you experience the relationship between these poles – chance and intention, algorithm and accident, rule and chaos – in your work as a creative coder?
The role of intentionality in generative art has really changed over the last year. Prior to ArtBlocks and fxhash, most generative artists would cherry pick from the outputs of the system they created. In a way, they would not present the system itself. Rather, they would present these different images that they had picked.
With the ArtBlocks approach, and the London Collection specifically, what I am creating is a singular system, which is also what I want to exhibit. With ArtBlocks, you don’t know what the outputs will be. You essentially have no control over them. The artist has to relinquish control to their system, even more than we usually do.
In a way, the intentional element is some- thing that certainly gets lost in ArtBlocks. I imagine the process with ArtBlocks almost like this randomly spinning wheel where you don’t know what you will get. As I mentioned earlier, the system-based dimension of all of this is very important to me. But I also wanted to have a curated element.
The artist knows the algorithm very well. And I want to find an exemplary image from the system I created. I am not talking about looking at 1,000 images to find the right one. One has to look at 100,000 images or more to find that one gem that really feels like the right output. What I am looking for here is that one specific image that lets the audience appreciate the process of the generative artist him or herself: an image in which all the different aspects of the system collide. Of course, this won’t yield an immediate understanding of the algorithm. But it can approximate the process of the generative artist: this collaboration with the algorithm that lets you find something in the diverse outputs of the system.
That’s fascinating. Could you say more about the properties of this one image that serves as the paradigm of the en- tire series?
If you start to look at a lot of outputs, you begin to scan them, and they start to look very similar. But if you look long enough, there might be one that looks radically different. That’s the kind of thing that I want to find. And it is something that is missing in most ArtBlocks projects, I think, because the range of outputs is limited to 100 or 1,000. To find this image, you need someone who knows the system well and who knows how far it can be pushed. In regards to the audience, I do not know whether selecting such an image will help them understand the algorithm, but it certainly helps them to understand the broadness of the system’s range.
Moreover, in the ArtBlocks projects, we deal with this construct of a singular entity since we associate the algorithm with a fixed file of JavaScript. But when you produce the work, you are constantly changing the algorithm. The algorithm is this thing that is in constant flux based on you seeing changes in the outputs. These tweaks in the code cannot be replicated on ArtBlocks. But the more we can show the audience that we are dealing with this variety of parameters, the more interesting generative art becomes. Generative art is really process-based in this specific way. The more you understand the process, the more interesting it becomes.
When did you decide to move your generative art on-chain?
The blockchain is a natural tool for generative art. With ArtBlocks, you can permanently “etch” the code onto the chain and thereby archive the algorithmic system. As I said, there is still too much of a focus on the singular image, but ArtBlocks, fxHash and similar really feel like we are placing more emphasis on the system and algorithm itself.
On the other hand, this archival function makes it difficult to encapsulate the constant changes that the algorithm would usually undergo in the creative process. In that way, it is an exercise for a generative artist. It puts you in a very specific mindset. The “long form” projects are the longest projects I ever created in terms of how much time I spend on them. And this is something that I heard from other artists as well. They spend months working on this singular algorithm.
You have explored different print techniques and light installations in your artistic practice. What do you think about the relationship between generative art and different media?
I think we are still struggling to find the right display solution. Since these outputs can be represented in various ways, there cannot be a singular solution, in my opinion. Some people want to explore the use of prints. Others open VR galleries. And this variety is due to the lack of a singular display solution, since there is no one way to visualize these systems.
For the sense of ownership, stewardship and storage, NFTs have brought a solution to generative artists. I think of something like Fidenza, let’s say. Before NFTs, there just wasn’t this concept of a singular, defining work of a generative artist in the public mind. Sure, there are important works like Vera Molnar’s “Letters from my Mother” (1984), but it hasn’t felt like these have entered mainstream study of art history and popular media and culture, in the same way that we have seen with Mondrian, Kusama, Magritte and other artists, and now we may be seeing in some small way in some NFT collections in how they are being regarded, analyzed, and talked about.
Could you say more about your thoughts on the lack of a singular display solution for generative art? What are the implications for the field?
The basic problem is that there is no clear output format for these algorithmic systems. This is because we do not have a way to immediately represent logical rules and parameters. We could do it as a text, as a diagram, as an animation ... Or even as a performance where someone is speaking the rules and the audience members are performing the actions. Most artists try to visualize these outputs, and this can be done in a variety of ways.
When it comes to exhibitions, I personally don’t like activations that are primarily just LCD screens, although it’s convenient and cost effective. To me, screens showing an image feel bland. If you just see a screen with an image, it is hard to appreciate the system that created the image. What I find much more interesting is to show the system- and process-based nature of the algorithm. If you have a print plotter, for instance, it allows you to feel closer to the system. Of course, you cannot really see the rules them- selves. But if you see the machine drawing a line, you can at least see the system producing the work in real time.
Generally, I want to find outputs that lend themselves more to the aspects of system and process. Performances are also an interesting idea, because there can be some- one executing the rules in the exhibition space. In this regard, I often think of Sol Lewitt’s work, specifically his wall drawings, which are essentially instructions to execute a drawing on the wall. Such artworks can showcase a system in an ideal way. With many of today’s generative art systems, however, the system is so complex that you cannot just have someone draw it on the wall. Hence you are in the trouble of finding the right output.
Based on what you just said, what are your thoughts about the future of exhibiting generative art in physical spaces? Will we see mostly white cubes with screens? Or are we moving towards more experimental and immersive installations?
I don’t think the display question is solved, and I am not sure if it ever will be. We just don’t have a way to encapsulate a system in the real world. Every project will explore different options to deal with this impossibility.
What is interesting about this problem is that it really gives us, as generative artists, the freedom to go into all of these interesting directions. Narrowing your practice to a single output media is not so crucially important. We have a system that can be outputted in all these different media and we are not bound to any specific medium.
If it is displayed well, then the exhibition is crucial for generative art. But for me, most current exhibitions fall short of that insofar as they simply put up a bunch of screens. Often, it’s literally a better experience to look at a website. But if the exhibition is done well, it can really help to reveal the system itself, which is a set of parameters, and not a set of images. This character of the system as black box of rules, parameters, and instructions is what I am trying to show in the London exhibition.
