The Interview | Operator

Prior to entering Web3, Operator were already renowned for their critical approach to technologies that invade user privacy and exploit digitally. Having exposed the surveillance state of Web2, this week they released their Art Blocks debut, Human Unreadable, to the acclaim of generative art collectors everywhere.

As the latest lot in their long-standing Privacy Collection, Human Unreadable adapts the duo’s experiential approach to the format of long-form generative art. In the process, it calls attention to Web3’s essential tension between transparency and privacy, while turning the blockchain into a space of eternal performance. Here, they explain how on-chain choreography can recuperate humanity to Alex Estorick.

Alex Estorick: How does your new Art Blocks project, Human Unreadable, extend the Privacy Collection?  ‍

Dejha Ti: The Privacy Collection focuses on the tension between privacy and transparency in blockchain technology, while also seeking to bring the human body into crypto art in a meaningful way. Human Unreadable does this by hiding the human body in plain sight — it’s on-chain and site-specific, but it also translates raw human expression into code and therefore obfuscates human movements in a way that is illegible to humans. The system is transparent yet the human body is still hidden — privacy and transparency.

The roots of the Privacy Collection, which comprises the Privacy Key 00, Attempts, Privacy Portraits, Unknown Sitter, and now Human Unreadable, lie in our works prior to Web3. First, our museum commission On View (2019), which was our way of analyzing selfie culture from the perspective of artists whose practice was being co-opted by experiential marketing, similar to Yayoi Kusama’s Infinity Rooms. On View coincided closely with the release of Shoshana Zuboff’s book, The Age of Surveillance Capitalism (2019). 

Our next project, I’d rather be in a dark silence than (2020), which won The Lumen Prize, is a signal-blocking trench coat that brought fashion, function, and conceptual art together to create a non-solution solution for how to combat the creep of extractive technologies into our daily lives. Both projects looked at privacy through a Web2 lens, so when we embarked on the Privacy Collection, we shifted our focus to how privacy functions within a Web3 context. 

We wanted to address the challenge and tension between privacy and transparency — being known for being unknown, hiding in transparent systems, using wallet addresses, personas, and profile pictures as a way to represent oneself and still be present, while maintaining a level of anonymity. 

The criteria for each lot of the Privacy Collection is as follows. One, it needs to address the tension between privacy and transparency. Two, it has to be site-specific to crypto culture. Of course, one tends to think of site-specificity in relation to physical space, but, for us, site-specificity in Web3 is more a mindset wherein the social dynamics and the technology itself are the four walls.

Finally, given that the human body is an essential component of our practice, we wanted to reinforce its importance to blockchain culture. Like the earlier lots in the Privacy Collection, Human Unreadable involves the materiality of glass, X-ray, and light together with the human body, but this time it does so in the context of long-form generative art. 

‍AE: I’m interested in the idea of crypto culture as a site and how that might have evolved over time. You’ve adopted a critique of art world institutions in the past but I wonder how you’ve adapted that strategy to Web3.‍

DT: When we created On View in 2019, it was kind of alien to the museum world. A number of traditional art world folks were skeptical as to whether it was really art or just spectacle or immersive entertainment. Once the project opened, we realized that some people weren’t taking it seriously as an art form perhaps because of the type of technology involved or the experiential format. We felt we needed to give it context. 

That same year, we did a talk at Christie’s Art + Tech Summit in Hong Kong and spoke at the first edition of CADAF in New York, where the Winklevoss brothers were buying CryptoPunks and DADA.nyc was inventing the “Invisible Economy,” encoding artist royalties into smart contracts for the first time. I think that the traditional art world had the same response to us as it did to them. Although we had made it into a museum, there was still distrust in our medium. The only people who seemed to be open and receptive to us were the crypto art folks, even though we weren’t creating art using blockchain at the time. 

Ania Catherine: We bonded with them instantly because of our shared experience of rejection and not being taken seriously because of how we were using technology and calling it art. 

‍In Web3, people want to be seen and they want their thoughts heard, but they don’t want to be known by their faces. Our Privacy Portraits — which anonymize the collector — were a response to the desire on crypto Twitter to be represented by something other than your face. That contrasts with Web2, which was about the selfie and about showing your face. 

‍AE: Your new generative project plays on the tension between encrypted identity and the transparency of blockchain technology. But presumably there is also a tension in working with a platform like Art Blocks that places limits on the projects on its platform. How were you able to sustain or reiterate your practice within the confines of the Art Blocks envelope?‍

DT: In order to ensure that the Privacy Collection is site-specific to crypto culture, we can’t ignore Art Blocks. Conceptually, it is sound to create a long-form generative artwork on a platform that has been integral to crypto art culture and the market. 

However, our only allegiance is to the concept of the work — we have no allegiance to any one medium. The question is which medium best serves the concept. 

The Art Blocks library demanded that we use p5.js, which meant that we had to reinvent how SVGs (Scalable Vector Graphics) work in order to draw body parts programmatically onto the canvas. It took a fair amount of complicated math in order to bring Bézier curves into the canvas. But every project has its own technical limitations and that’s nothing new. ‍

AC: The experience of developing Human Unreadable also expanded our practice immensely, which I find interesting because it involved the most specific format restrictions of anything we’ve ever made. We also created tools that can now be used by other artists. It’s a cliché that creative constraints can actually expand what you do, but in this case the technical constraints helped to produce something massive and nuanced that doesn’t feel constrained at all. 

‍DT: The project also involved a vast amount of motion data. Blockchain technology is great for many things, like keeping a transparent ledger, but it’s not really intended for high-volume data storage. 

Even if we had $200,000 to upload a vast library of high fidelity motion data on-chain, that doesn’t really feel in the spirit of blockchain. It wouldn’t make much sense and it wouldn’t be respecting the medium. 

Instead, we had to embark on the process of both compressing the data and getting it into a usable format of reasonable size. That process alone was one of the biggest challenges we faced on this project. 

‍AE: Other performance artists have taken a pre-existing performance and fixed it on the blockchain as a means of selling it as an NFT. It feels like you’re inverting that logic here given that the performance is emerging from the blockchain itself.‍

AC: With the “Privacy Collection” we always intended the performances to get funkier over time. With Attempts (2022), we captured a performance that was then sold subsequently. With the Privacy Portraits (2022), the artwork was created through the performance, so it actually took place in the present. In the case of Human Unreadable, the mint generates the performance, which then takes place subsequently. So the performance component has evolved over time.  ‍

DT: On Art Blocks, there’s always been a token hash that produces a reliable final output that is unique. This is true of Squiggles and every other project on the platform. As a result, different collectors never own the same work or experience. However, we decided to add another step, taking the token hash and creating what we call a “choreographic hash.” That choreographic hash fed all of the visual parameters that drove the final visual output. 

But more than simply an input, the choreographic hash is also a movement score in itself — a unique choreographic sequence. That means that every collector not only has the PNG stored on IPFS but also an underlying choreographic score that is unique to them. 

‍AE: Web2 is premised on the capture of user data and the construction of “data bodies” that are then packaged for sale, usually without the involvement or consent of the user. Your new project seems to posit an alternative situation where the body emerges as a kind of hybrid, emergent form out of a generative system. How do you understand it?‍

DT: As we developed Human Unreadable, we sought to ensure the translation and abstraction of Ania’s original physical movements into a form appropriate to the medium in question. As a consequence, the final performance involves a degree of translation and is not a perfect representation of the on-chain choreographic score. 

The work involves five moments of translation: Ania’s original creation of single moves that make up the Human Unreadable movement library; the registration of that abbreviated choreography on-chain; the translation of that into a human readable on-chain choreographic score; and then the choreographers’ and, ultimately, the performers’ own interpretations of that score in the flesh.‍

AC: When you were talking about the emergence of the human, I immediately thought of each person performing a different sequence based on the unique outputs of the generative system. 

When all of the outputs are performed together in one shared space it becomes a highly dynamic, sometimes almost violent, performance. 

As part of our JPG x MOCA performance, we tuned the model in front of a live audience. We ran the model, which was then projected, and I did each of the individual movements myself while wearing a motion capture suit. I would then teach the performers each sequence which they then performed together. The model determines whether a particular sequence is performed as a solo, duet, trio, or in a larger group.

‍DT: The piece minted on Art Blocks is a still rather than a moving image. If no one told you, you wouldn’t suspect that the work had been produced via choreography. Indeed, if you look at the list of features, which can be used to determine rarity in a generative art collection, you won’t find any visual features at all. Instead, the rarity of each output depends on its underlying choreography, which is not visible in the mint. People will say: “I have the one with the most unique moves, or I have the one with the solo, or I have the one with the most repetition, or the one where the performance is facing west.” We also added vulnerability as a feature, so you’ll have a vulnerability score, although that isn’t tied to an output’s particular visual characteristics. 

Usually, when you reveal a work on Art Blocks, the output is the end of the artwork. In Human Unreadable, that is the beginning of the work — the moment when the slow recovery of the human begins. But it’s also the moment at which the human component is entirely unreadable. 

The second moment of human recovery is when you go to the Operator site in the weeks following the mint and connect your wallet. At that point, collectors will receive a secondary NFT that is forever bound to the primary Art Blocks NFT. It is that secondary NFT that reveals the work’s underlying choreographic score. 

Of course, the secondary NFT is also an additional form of decentralized storage. We know that Right Click Save cares about NFT security and we do too. This work basically backs up the collector’s score twice and potentially a third time when you consider that the choreography will then be taught to dancers, which is how choreography has been passed down historically. The final recovery of the human takes place during the live performance when the sequences behind the first 100 collected mints are performed in physical space. At that point, you encounter long-form generative art that can’t be right click saved.

‍AE: What are the implications of this “recovery of the human” from a political, ethical, and non-art perspective? Is there a way one could read the process as not so much a recovery of the human but as an emergence of a posthuman?‍

AC: Right now, I feel like disembodiment is the norm and our desire to recover the human is really about redirecting movement, capital, and cultural momentum. I believe in the power of performance to transform the current mode of hyper-speculation into a form of utility whereby collectors get to go to a major museum to see something they collected performed in front of them. 

The transition from a natively digital experience to a person-to-person exchange is a victory. There’s no other way that I can see most of these collectors ending up in that position. Our work is a way to facilitate that connection — it’s a slow recovery of the human and a way to remind people of the power of human exchange through performance. ‍

‍DT: While on-chain generative choreography might be considered novel, ultimately we want Human Unreadable to reinforce the role of women as central to the history of digital art, while moving away from modernism towards embodiment, chaos, and sensuality.