Interdependent Variables: Remotely Designing Tactile Graphics for an Accessible Workflow
Lilian de Greef, Dominik Moritz, Cynthia Bennett · 2021 · Proceedings of the 23rd International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) · doi:10.1145/3441852.3476468
Summary
This experience report documents how a team of blind and sighted colleagues at Apple created an accessible collaborative workflow for a data visualization research project. Rather than focusing on making end products accessible, the paper addresses the less-explored challenge of making the research process itself accessible — ensuring that a blind team member (Cynthia Bennett, a braille reader and tactile graphics user) could participate equally in data-driven discussions. The team developed both handmade and machine-embossed tactile graphics to serve as shared data representations that all team members could reference during remote collaboration during the COVID-19 pandemic. For handmade graphics, the sighted colleague (Lilian) taught herself tactile graphic design through community workshops, studying embosser manufacturer samples non-visually, and consulting with Cynthia on preferences such as Grade II braille and corner-cut orientation markers. She created material reference sheets cataloguing different textures and their tactile qualities, then crafted charts using puff paint, textured tapes, and other 3D materials with both ink and braille labels. For machine-embossed graphics, the team used a ViewPlus Columbia 2 embosser with TactileView software, developing a tight feedback loop where Lilian digitally created graphics, Cynthia provided detailed tactile feedback, and they iterated together. The graphics were physically mailed between team members since all work was remote.
Key findings
The team found that tactile graphics transformed their collaboration, becoming the basis for discussion and enabling "aha" moments that verbal descriptions, numeric tables, or alt text could not achieve. For example, after hearing a line chart described as "volatile," examining the tactile version revealed exactly how volatile the data was — nuance that words alone could not convey. Verbal descriptions inevitably had gaps: a sighted interpreter might not realize a detail was crucial, or their explanation might be complicated and long. Handmade tactile graphics sometimes offered higher fidelity than embossed ones, particularly for chart types benefiting from varied textures or three-dimensional materials like heat maps. However, embossed graphics allowed faster iteration and easier production of multiple copies for additional BLV collaborators. The team applied visual chart design best practices to tactile form — for instance, de-emphasizing gridlines (making them faint relative to data) and translating color saliency into texture saliency. A critical broader lesson was that making accessibility a shared team responsibility — rather than placing the burden solely on the disabled team member — required explicit investment in labor, time, materials, and institutional recognition. The team emphasized that this access work must not remain "invisible" or be treated as an add-on, but must be properly resourced and acknowledged.
Relevance
This paper shifts the accessibility conversation from products to processes, arguing that inclusive outcomes require inclusive workflows. For teams that include blind or low vision members, the practical lessons are immediately applicable: create tactile reference sheets before starting chart production, use both handmade and embossed approaches depending on fidelity needs and iteration speed, organize tactile graphics in labeled binders for easy reference, and use shared spreadsheets so all team members can find the same graphic. The framing of accessibility as "interdependent" — co-created by all team members rather than accommodated for one — challenges the dominant model where accessibility is siloed into a separate office or treated as an individual burden. For organizations, the key takeaway is that accessible collaboration requires institutional support: budget for materials and equipment, time allocation for access work, and recognition that this labor contributes to better outcomes for the entire team. The paper also demonstrates that current digital accessibility tools (alt text, screen reader descriptions) are insufficient substitutes for tactile graphics when doing substantive data analysis work.
Tags: tactile graphics · blind and low vision · data visualization · collaboration · accessible workflow · remote work · interdependence · inclusive design