Domestic Data Streamers is a Barcelona-based research and design studio working at the intersection of data, community, art and social transformation. Founded in 2013, the studio has developed hundreds of installations, exhibitions and participatory projects that challenge the way data is usually understood, shifting it from abstract information into something that can be physically experienced, emotionally processed and collectively negotiated.
Rather than visualising data in conventional ways, their work often attempts to stage encounters: translating datasets into materials, gestures, environments and situations where audiences are not simply viewers, but participants. Projects such as Uncoded Gestures (2025) reflect a broader investigation into how statistical, technological and historical structures can be made present without being reduced to explanation.
I spoke with Pau Aleikum Garcia, Founder and Director of Domestic Data Streamers, about data as a “ghost,” cognitive openness, institutional anxiety and the fragile balance between ambiguity and clarity in contemporary exhibition-making.
Domestic Data Streamers, Uncoded Gestures, 2025. Installation view, Escoleta Building, Palo Alto, Barcelona. Courtesy Domestic Data Streamers.
Filippo Lorenzin: Your work often sits somewhere between data, narrative, and experience. When you start a project, do you think of data as a material to shape, a language to translate, or something else entirely?
Pau Aleikum Garcia: Data sometimes feels like a spirit, and I believe it literally. You can understand the number 10,000, but you cannot see it. It lives entirely in your head. Numbers are abstract beliefs we collectively agree to trust, which, if you think about it, makes statistics weirdly theological.
So when we start a project, we are not really “shaping a material” or “translating a language.” We are trying to make something invisible become present. Physical. Confrontational, even. Probably the biggest contradiction we face is that data can be intangible, but the emotions and impact that data represents are not. Behind an unemployment percentage, there are families who cannot make it to the end of the month. Behind a gender violence statistic, there are 730 real hours of violence. Our job is to find the bridge between the number and the person.
Filippo Lorenzin: And is that bridge always emotional, or can it also be spatial, material, even uncomfortable?
Pau Aleikum Garcia: So if I had to pick a word, I would say data is a ghost. And we build the machines that make ghosts visible. Wittgenstein wrote that the limits of my language are the limits of my world. We take that quite seriously. If the only language available for a dataset is a bar chart, then the world that dataset describes stays very small. We are trying to expand the number of languages available, ceramics, sand, thread, sound, and space, so the understanding can be wider. Whether that makes data “a material” or “a language” I honestly do not know. It is probably both at different stages of the process, and neither when you zoom out far enough.
Domestic Data Streamers, Uncoded Gestures, 2025. Audience interacting with the main installation, Palo Alto, Barcelona. Courtesy Domestic Data Streamers.
Filippo Lorenzin: Many data-driven projects risk becoming either too abstract or overly didactic. How do you navigate the balance between intellectual rigour and emotional or intuitive engagement?
Pau Aleikum Garcia: The question we have become most obsessed with is not really about the balance between rigour and emotion. It is about something that comes before both, cognitive openness. How do you get a person into a state where they are actually willing to reconsider what they think they know? There is a researcher, Claudia Cano, from the University of Advanced Studies of Social Sciences in Paris, who came to present at our studio and helped us articulate something we had been doing for years without fully understanding the mechanism. Her work sits at the intersection of aesthetic and epistemic experience, which sounds heavy, but the core idea is surprisingly simple. Every organism builds a predictive model of the world. We walk around with these internal simulations running constantly, expecting things to behave as they have before. The kind of learning that changes behaviour happens when something breaks that model, when what you expected does not match what you encounter.
Aesthetic experience does this particularly well because it operates outside utilitarian logic. When you engage with something aesthetically, a sound, a texture, a space, you approach it for its own sake, without the defensive filters you would bring to a political argument or a classroom lecture. Several research studies show that this kind of intrinsic engagement produces deeper, more lasting learning than extrinsic motivation. You do not retain information because you were told to, you retain it because you wanted to understand it. So what we actually design are the conditions for curiosity.
Filippo Lorenzin: That’s interesting, because it shifts the role of the artist or designer from explaining something to setting up a situation. Is that how you see it?
Pau Aleikum Garcia: Loewenstein’s information gap theory explains that curiosity arises when you become aware that you are missing something, not when you are given the answer, but when you sense there is a gap. We create gaps with that goal, then it becomes easier to bring rigour and create a space where an audience can decide to continue exploring or simply take the surface message.
Filippo Lorenzin: There is a long history of artists working with systems, archives, and information. Where do you see your practice positioned in relation to that lineage?
Pau Aleikum Garcia: The honest answer is that we arrived at this work sideways, and most of our references reflect that. Before DDS existed, I was working on memory reconstruction projects in Berlin and Norway, with archaeologists, architects, and historians, trying to understand how communities remember conflict and how physical spaces carry that memory. That work was about the same fundamental problem, how do you make invisible things present, how do you give weight to what has been lost or deliberately hidden.
When we started the studio in 2013, the people who most shaped our thinking were practitioners who treated data not as an end but as a bridge to something human. Jaime Serra’s work was important early on, the way he brought emotional and biographical texture into infographics at a time when the field was almost entirely concerned with clarity and efficiency. It was similar to the approach of Nicholas Felton and his reports, showing how the act of quantifying personal experience could itself be a form of self knowledge, and also a precursor to reflections on surveillance and how information technology in the wrong hands can be very harmful. Jonathan Harris was also a very important reference. His way of talking about stories using both digital and physical media seamlessly was really impressive to us. On the formal and spatial side, we have always watched David Bowen and Random International closely, especially in how they created environments where the audience’s body is implicated in the work.
What we consciously try to distance ourselves from is a particular tendency within data art, the fetishisation of the dataset itself.
Domestic Data Streamers, The Citizens’ Office of Synthetic Memories, 2024. Installation view. Courtesy Domestic Data Streamers.
Filippo Lorenzin: Do you think this is also a problem of how data art has been historically exhibited and commissioned?
Pau Aleikum Garcia: There is a tradition where the aesthetic is the complexity, beauty, and elaborate visualisations that end up keeping data locked in its abstract form. The other thing we try to resist is the single author model. Chantal Mouffe’s idea of agonistic pluralism, that productive conflict rather than consensus drives democratic discourse, sits closer to how we think about exhibitions than any art world framework. We design for disagreement, for complexity that can be felt before it gets rationalised away. We always try to ensure that the audience completes the work, and if they do not, we have just made a static sculpture, no disrespect to sculptures, but that is not what we are trying to do.
Filippo Lorenzin: I am interested in process as much as outcome. Could you walk me through a recent project, from the moment you encounter or generate a dataset to the point where it becomes a public-facing work?
Pau Aleikum Garcia: I will use Uncoded Gestures, which we opened in April 2025 at the Escoleta building in Palo Alto, Barcelona. It started from two things converging. First, companies like Veesion are training AI on CCTV footage to detect shoplifting through body language, and they have flagged over 100,000 “suspicious” gestures. For these companies, our bodies have become texts that can be continuously read, stored and policed by algorithms. Second, Andrea De Jorio, the antiquarian who in 1832 noticed that gestures depicted on ancient Greek vases looked remarkably similar to everyday body language in contemporary Neapolitan markets. After two thousand years, the same raised arm, the same tilt of the head. Gestures outlast empires, languages, and religions. In our attempt to use AI for something other than automation and optimisation, we started to ask how the same tools designed to surveil bodies could instead connect people to the past and to history.
We started by accessing historical photograph archives from Hospitalet and Poblenou, rich collections from neighbourhoods near us. We built a prototype using pose detection models that could match a visitor’s live gestures to archival images. You stand in front of a camera, move naturally, and the system finds a historical photograph where someone, a striker in 1977, a dancer in 2003, a kid playing in 1965, held the same posture you are holding now.
The technical prototype was one layer.
Filippo Lorenzin: So the technology is not the centre of the work, but one of the conditions that allows the encounter to happen?
Pau Aleikum Garcia: A lot of the actual design work was everything around it, the spatial layout, the questions we posed on the walls, what defines you more, your gestures or your words, and the tension between surveillance technology and something closer to personal archaeology.
What we did not fully anticipate was how people responded to the idea that their bodies carry inherited scripts, that a shrug or a nod between strangers might be something passed down through generations without anyone consciously deciding to pass it down. The conversation shifted from “look what AI can do” to “what does my body remember that I do not?”
Domestic Data Streamers, The Time Keeper, 2016. Installation developed with Sónar Festival and Spotify. Courtesy Domestic Data Streamers.
Filippo Lorenzin: You describe data as a kind of “ghost,” something that becomes present only through the systems you build. How does this translate when the work is exhibited? Do institutions tend to reinforce that invisibility by framing data as information, or can they support a more ambiguous, affective encounter?
Pau Aleikum Garcia: Honestly, this is something we are still figuring out, exhibition by exhibition. The short answer is that institutions can do both, and often do both at the same time. The longer answer is that the institution is never just a container for the work, it is part of the machinery.
Filippo Lorenzin: This is a very important point. Does it mean that exhibition-making becomes part of the work itself?
Pau Aleikum Garcia: When institutions reach for “data as information,” I do not think it comes from bad faith. It comes from anxiety. There is a real structural pressure to make sure no visitor feels excluded, no funder feels confused, no school group leaves without a learning outcome, and no journalist gets bored at the press preview. So the tendency is to translate the work into a paragraph that begins with “This piece explores” and ends with “inviting reflection on our digital age.” Once that paragraph is on the wall, the work is already half buried under it. The ghost becomes a footnote.
But some institutions understand that ambiguity is not the enemy of access, it can be the way in. When an institution trusts the audience to sit with confusion for a few minutes, the affective encounter has space to actually happen. We want to get answers from the body before the brain does it.
What we have learned is to treat the framing layer with the same care as the piece itself. We rewrite wall texts, we argue for fewer words, we push for a question instead of a description. Sometimes it works, sometimes it does not, but we always try.
Filippo Lorenzin: You mentioned designing for cognitive openness and creating conditions for curiosity rather than delivering answers. In an institutional setting, where there’s often pressure to communicate clearly and efficiently, have you encountered resistance to this approach?
Pau Aleikum Garcia: Constantly. And not because the people pushing back are wrong. An institution has obligations we do not have when we are alone in the studio. They have school groups arriving on Tuesday morning and a press team that wants a one line summary by Friday. The whole apparatus of any museum or cultural institution is calibrated to communicate, to make sure no one feels lost or excluded, and that no one writes a complaint email. Inside that pressure system, the idea of “designing a gap of information” can sound either pretentious or irresponsible.
The most common moment of friction is around what I would call the pre-emptive answer. Two weeks before opening, someone, sometimes a curator, more often someone in education or communications, starts asking what the audience will “take away.” It is a fair question, but it tends to arrive too early in the process. And the moment you commit to a takeaway, you are already closing off any alternative interpretation of the work. You move from creating the conditions for curiosity to delivering a message.
Filippo Lorenzin: Is that pressure stronger today, now that institutions are expected to be accessible, measurable, and accountable at the same time?
Pau Aleikum Garcia: So what we do is try to sit with educators long enough to find a shared language. Most of the educators I have met are not the enemy of ambiguity. They are often the most sophisticated readers in the building. What they are afraid of is a visitor leaving without anything to hold onto, and I share that fear. So the negotiation is rarely “ambiguous vs. clear,” but rather about where the clarity lives. Sometimes the wall text can hold it, or sometimes it is an educator’s voice in conversation. Sometimes a single object in the room becomes the anchor, and the rest of the space is allowed to drift. We have had to learn to distribute clarity rather than concentrate it.