Megan Hofmann is a leading researcher in Automated Machine Knitting and focuses on developing infrastructure for computational knitting techniques.
For non-technical domain experts and designers, creating designs that meet domain-specific goals can be a substantial challenge. This presents an opportunity to create specialized tools that produce optimized designs in the domain. However, implementing domain-specific optimization methods requires a rare programming and domain expertise combination. Creating flexible design tools with re-configurable optimizers that can tackle various problems in a domain requires even more domain and programming expertise.
Digital fabrication technologies, everything from consumer 3D printers and laser cutters to industrial knitting machines, are changing how we build the world around us and, more importantly, who gets to build that world. In the context of healthcare, 3D printing is changing how we create assistive and medical devices at the point of care and proved to be an essential tool for producing PPE during the COVID-19 pandemic. Despite these recent successes, a more careful examination of how healthcare professionals adopt digital fabrication technologies reveals a mismatch between what design tools are made to do and what clinicians want to do. Dr. Hofmann’s research reverses our expectations of who uses design tools and who builds them by bringing clinical domain experts into the process of building digital fabrication systems.
She has conducted extensive work on digital fabrication in various healthcare contexts, teaching us how medical domain experts approach fabrication challenges.
Tactile maps can help people who are blind or have low vision navigate and familiarize themselves with unfamiliar locations. Ideally, tactile maps are created by considering an individual's unique needs and abilities because of their limited space for representation. However, significant customization is not supported by existing tools for generating tactile maps. We present the Maptimizer system which generates tactile maps that are customized to a user's preferences and requirements, while making simplified and easy to read tactile maps. Maptimizer uses a two stage optimization process to pair representations with geographic information and tune those representations to present that information more clearly. In a user study with six blind/low-vision participants, Maptimizer helped participants more successfully and efficiently identify locations of interest in unknown areas. These results demonstrate the utility of optimization techniques and generative design in complex accessibility domains that require significant customization by the end user.
Accessibility research and disability studies are intertwined fields focused on, respectively, building a world more inclusive of people with disability and understanding and elevating the lived experiences of disabled people. Accessibility research tends to focus on creating technology related to impairment, while disability studies focuses on understanding disability and advocating against ableist systems. Our paper presents a reflexive analysis of the experiences of three accessibility researchers and one disability studies scholar. We focus on moments when our disability was misunderstood and causes such as expecting clearly defined impairments. We derive three themes: ableism in research, oversimplification of disability, and human relationships around disability. From these themes, we suggest paths toward more strongly integrating disability studies perspectives and disabled people into accessibility research.
With the increasing popularity of consumer-grade 3D printing, many people are creating, and even more using, objects shared on sites such as Thingiverse. However, our formative study of 962 Thingiverse models shows a lack of re-use of models, perhaps due to the advanced skills needed for 3D modeling. To improve reusability, our framework (PARTs) allows modelers to graphically specify design intent through geometry with embedded functionality. PARTs includes a GUI, scripting API and exemplar library of assertions which test design expectations and integrators
which act on intent to create geometry. PARTs provides a way to integrate advanced, model specific functionality into designs, so that they can be re-used and extended, without programming. In two workshops, we show that PARTs helps users create 3D printable models, and modify existing models more easily than with a standard tool.
