Peopling.studio

a Q&A with Galo canizares

To start: can you give us a quick overview of what peopling.studio is about, and why you chose to develop it?

Peopling.studio is part of a larger research project on virtual embodiment and identity. It parallels a recent essay published in the Journal of Architectural Education 74-2 that surveyed processes of othering in virtual environments. The essay called attention to biases in the ways humans are represented in software. It’s not difficult to notice these phenomena: 3D avatar design tools hyper-sexualize both male and female forms, default figures are always white, lack of representation of different ability levels, etc.

The premise was to provide free 3D scenes that showcased not universal, averaged humans, but characters that could tell stories: scale figures as action-figures. It’s aimed primarily at students who want easily usable 3D objects that are neither overly cartoonish nor uncannily realistic like 3D-scanned people. More importantly, it reverses the typical defaults found in software and assets websites, such as skin color and ability.

Image courtesy of Galo Canizares and office ca.

What can you tell us about your design process for these figures? Were you working from “real person” models? How did you decide on them, or what were your criteria for inclusion of the characters that comprise peopling.studio?

These particular figures begin with a base mesh exported out of a software that’s unfortunately no longer in production, Mixamo Fuse. This geometry is then tweaked in Blender (the earlier versions were made in 3ds Max) to customize its features, and then rigged using an auto-rigger, which gives it a reconfigurable skeleton. In general, the technical process is pretty messy. I’ve been working with these tools for a few years, and they still require a lot of tinkering to get right. Importing/exporting meshes, simplifying geometries, rebuilding parts completely...it’s not super streamlined.

The main criteria for the characters is to fill a gap in representation. I think: what kinds of scale figures have I not seen? Who is not being represented? And go from there. Of course, there are a lot of answers to those questions, but that’s why the project can keep going forever. I suppose another set of criteria is to directly oppose what you see in popular 3D environments. There is a desire to tone down the sexuality of 3D figures (which is sometimes hard to do because default male and female body meshes are often super exaggerated in this software). Sometimes scale figures don’t need explicit sex-related features or big muscles at all. It’s a difficult process because it inevitably reveals biases, my own or the tool’s. That’s why I consider the project a form of storytelling, and why I emphasize that what peopling.studio provides are not “objects,” but rather “scenes.”

One thing we really appreciate about this project is its projective, actionable quality. Where a lot of the conversation about diversity and its representation in architecture remains just that - conversation (which is not to belittle its value) - this project moves beyond the “critical” and towards the projective or speculative. Were you thinking along those lines at all? And, relatedly, are you aware of other projects that embody this same actionability? (We saw the Open Source Afro Hair Library you link to - excited about that!)

Yes, the project came out of a desire for action. I think architects rarely think of themselves as activists, but we do possess a unique set of skills that can effect change, given the right context. Peopling.studio was a means to practice what I preach. I have been writing about these issues since about 2018, but eventually (and thanks to being stuck at home during quarantine) I was compelled to put my knowledge of 3D modeling to good use.

There are some fantastic projects out there that tackle similar themes. Nonscandinavia.com, black-img.com, and peopleparty.app are just some incredibly useful and ethically responsible peopling initiatives. Dimensions.com is somewhat tangential, but an amazing resource, too.

We’re curious about how you incorporate scale figures (“peopling”) in your (non-peopling.studio) work. Do you have a theory/strategy in regards to when/where/how you use scale figures? Or, more generally, the role of the scale figure and its representation (generic vs. specific, abstract vs photoreal etc.) in architectural imagery?

My partner and I have very strong opinions on the role of scale figures in representation. I wouldn’t necessarily call it a theory, but in general our position is that scale figures should tell good stories and further the narrative of the project. Short of giving them names, we regard our scale figures in our design projects as characters, each contributing to an understanding of the project in some way (beyond the function of scale).

In our design work, we also use photo-realistic techniques in nonstandard ways to call attention to the virtuality of the medium. We’re super interested in the uncanny feel of The Sims, the paintings of Hieronymus Bosch, and the illustrations of Max Guther. So while we may use the same visualization tools as, say, Luxigon, we like to put a quirky spin on renderings and drawings. We think of our modes of representation as either re-enactments, scenarios, or simulations, all narratives featuring people in one way or another. For a general example of this type of imagery, I might point out the illustrations in my book Digital Fabrications: Designer Stories for a Software-Based Planet, which are essentially re-stagings of several historical photographs in 3D made to avoid purchasing image rights.

We’re wondering about your software workflows. Especially what we might read as a preference for Blender, and a potential extrapolation to free/open source software as an emancipatory tool?

Free is always good (that’s why Skalgubbar got so popular!). But I grew up in the days where you really had to work hard to get free things off the internet, legally or otherwise. As an educator, I’m committed to making design tools more accessible to a wider range of students. I refuse to teach proprietary software unless it is (1) provided by the school, or (2) all of the students already own it. That is to say, I will never ask students to buy software for a class or workshop I’m teaching. Luckily, today’s free software is becoming increasingly powerful; the latest version of Blender is now, in my opinion, comparable to 3ds Max or Maya, which I’ve used for years. My license for Autodesk expired last year, and so I moved to Blender for most 3D work and I do not regret it. I use Rhino as well, and am happy they offer an affordable educational version that is not a subscription model (my current institution also generously provides our students with Rhino licenses).

Free and open-source software is quite liberating. In addition to Blender, I frequently use Processing/p5, Unity, three.js, and have dabbled with Unreal Engine. What I think is most important about these tools is that they invite you to tinker and get familiar with programming. Code is perhaps the most emancipatory tool at the moment.