Between late 2018 and late 2024, I have been working on my dissertation at Ulm University, defending it on January 22nd, 2025. The dissertation is based on 8 publications in total: Ubiquitous Fabrication, Immersive Sampling, Mix&Match, ShapeFindAR, pARam, BrickStARt, DungeonMaker, and Ephemeral Fabrication. All projects explore futures of personal digital fabrication by developing design tools, interaction approaches, and proposing alternative lenses on personal fabrication. The document itself provides an overarching structure for all 8 papers to be situated in (expanding upon Ubiquitous Fabrication), but similarly aims to provide additional detail, context, and motivation for all projects.

Links

The dissertation is available through the Open Access Repository of Ulm University and Ulm University of Applied Sciences (OPARU): 10.18725/OPARU-55953. Note that the document there had to be compressed significantly. Contact me for an uncompressed version, if needed. The papers in the appendix are affected the most, but they are available as preprints or are open access.

Abstract

Personal computing has transformed society by providing industry-grade computing capabilities to end-users: complex, expensive, and expert-only mainframes turned into ubiquitous devices essential to a vast user audience. In the context of Personal Fabrication, users recently gained access to industry-level tools and processes (e.g., 3D printing) to design and manufacture various physical artifacts. While computing evolved its interaction paradigms towards more approachable interfaces (i.e., from command lines to graphical user interfaces), Personal Fabrication still enforces a paradigm of ex-situ modeling, fabrication, and iteration until the user achieves a satisfying result. Devices like 3D printers or laser cutters are increasingly easy to use and affordable, yet designing for manufacturing remains a highly complex task: users have to tie together a multitude of disciplines, ranging from engineering over material science to design. Subsequently, Personal Fabrication presents a powerful opportunity, yet it predominantly remains employed by hobbyists and enthusiasts.

This thesis aims to resolve the tensions between potentially highly expressive fabrication (i.e., physical output) devices and the complex design tools (i.e., digital input) they demand—how do we reconcile low-effort interactions with a potentially unlimited physical output space? To answer this question, the thesis first presents a theoretical framework, which encompasses design tools for Personal Fabrication. These tools can be arranged in a process model that covers design, fabrication, and the use of the outcome. The framework focuses on the design step and classifies the effort and expressivity of design tools to formalize three main paradigms of physical artifact design: “modeling”, “remixing”, and “getting”. Most established design tools focus on the “modeling” paradigm, where users routinely re-design existing objects on their way to their finished design. This approach demands, often unnecessary, effort. Such design steps are also usually disconnected from the location of use (ex-situ). Users regularly conduct design steps on workstations like desktop computers to design objects for the physical world. This disconnect requires users to transfer requirements between design and usage contexts. The core of the thesis, therefore, focuses on a) partially or b) entirely omitting or c) fundamentally altering steps of modeling. These changes are enabled by relying on outsourced design effort and in-situ design tools to foster low-effort interactions. This approach manifests in the development and evaluation of several prototype systems that deliberately rethink the notion of design for Personal Fabrication.

Derived from these works, the thesis provides insights into the design of design tools for Personal Fabrication. By deliberately omitting or reconsidering workflow steps (e.g., 3D modeling), the presented prototype systems reduce the effort needed to design an artifact for manufacturing. This effort reduction is based on design principles contrasting established core paradigms (modeling) and towards in-situ workflows leveraging users’ unique physical contexts. Specifically, instead of starting “from scratch”, users may remix existing models, tune parametric designs, or merely retrieve their desired artifacts. These approaches move processes in Personal Fabrication away from complex but powerful industrial CAD (computer-aided design) systems without constraining their relevant expressivity. Similarly, instead of relegating design processes to a separate workstation, users may conduct search, remix, and preview procedures in-situ at the location of use for the future artifact. In-situ interaction simplifies the transfer of requirements from the physical environment. The paradigms of remixing or getting, blended with in-situ interaction through extended reality, are the core pillars of the thesis.

The thesis concludes by reflecting on the broader vision of “Ubiquitous Personal Fabrication”, where the notion of digitally-supported craft and manufacturing is woven into everyday life, akin to digital content creation enabled through ubiquitous computing. In this context, we may see a future where anyone can create highly personalized artifacts that suit their unique contextual, aesthetic, and functional needs without precisely defining every single detail of the artifact, empowering a broad range of users, regardless of their motivation and proficiency, to engage in design activities.

Figures

For the dissertation, I created a set of additional figures to illustrate aspects of the projects, or concepts surrounding design and personal fabrication. They have not appeared in any publications yet, and are presented here as a bonus.

Designing in Bits / Designing in Atoms.

An in initial reflection involved considering the dichotomy of ‘‘bits’’ and ‘‘atoms’’, as labels for digital or physical artifacts respectively. Design activities can involve users manipulating bits or atoms to manifest their desired outcome. Digital Fabrication lets users design artifacts digitally first (‘‘in bits’’) and receive a physical outcome (‘‘atoms’’).

✖️

Digital Fabrication allows digital designs to transition from bits to atoms.

Notably, designing in bits and designing in atoms are separate streams, but have converged within the notion of Digital Fabrication, an may later remain intertwined in what we know as Sutherland’s ‘‘Ultimate Display’’, as introduced in 1967.

✖️

Practices of designing in bits or designing in atoms converge, in a distant future, in the Ultimate Display. However, they are already intertwined in Digital Fabrication.

Process.

A few reflections on the ‘‘process’’ and individual tasks associated with personal fabrication. The starting point was the depiction of Interactive Fabrication by Willis et al., presented at ACM TEI 2011.

✖️

Digital Fabrication usually involves CAD (Computer-Aided Design), CAM (Computer-Aided Manufacturing), and yields a physical outcome.

✖️

The CAD → CAM → Outcome pipeline can be simplified into steps of Design, Materialization, and Use. Materialization is meant to depict a more generic lens on fabrication, including fabrication steps delegated to services and commercial artifact acquisition (e.g., stores) as feasible ways to conduct this step. A focus on various processes and the craft itself is suitable for technology-enthusiasts and hobbyists, whereas a focus on outcomes is more suitable for a broader audience of consumers.

✖️

All steps from design, materialization, and usage can be further decomposed: they involve different activities that users may either engage in, or omit. The more steps a user can reasonably omit, the lower the effort they have to invest to get to their envisioned outcome.

✖️

Proposed alternative approach, compared to a potential depiction of the status quo. Intent and outcome are usually situated in a physical context and are bridged through a set of tools.

Users / Motivations.

Who engages with ‘‘acquiring physical artifacts’’ and what could be their motivations? What are platforms and tools that users may rely on and how large are the individual populations that we are talking about? These questions—I believe—are crucial to consider.

✖️

Rough, schematic Venn-diagram of user motivations for personal fabrication and practices or services that may cater to these requirements.

BibTeX

@phdthesis{stemasov-phd-thesis-2025,
    title = {Enabling Ubiquitous Personal Fabrication: Low-Effort and Expressive In-Situ Interactions Using Extended Reality},
    author = {Stemasov, Evgeny},
    publisher = {Ulm University},
    doi = {10.18725/OPARU-55953},
    pages = {1--398},
    year = {2025}
}

📋 Copy BibTeX
📥 Download BibTeX