Tactile Vega-Lite: Rapidly Prototyping Tactile Charts with Smart Defaults
Mengzhu (Katie) Chen, Isabella Pedraza Pineros, Arvind Satyanarayan, Jonathan Zong · 2025
Read Tactile Vega-Lite: Rapidly Prototyping Tactile Charts with Smart Defaults by Mengzhu (Katie) Chen, Isabella Pedraza Pineros, Arvind Satyanarayan, Jonathan Zong (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-00149.txt.
Introduction
Mengzhu Chen, Isabella Pedraza Pineros, Arvind Satyanarayan, and Jonathan Zong build a tactile chart tool in a 2025 paper. Tactile charts let blind and low vision readers feel data, but they are hard to design. New designers face dense guidelines, and experts face slow, scattered workflows. Interviews with expert designers shape the tool.
Tactile Vega-Lite extends a known chart grammar with tactile defaults drawn from those guidelines. New users get compliant charts fast, and experts can override any default. A study with twelve creators shows faster cycles, more consistency, and automatic braille and spacing work, the authors report. Expert critique adds new design lessons.
Catalog record
| CREATOR | Mengzhu (Katie) Chen, Isabella Pedraza Pineros, Arvind Satyanarayan, Jonathan Zong |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-00149
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Tactile Vega-Lite: Rapidly Prototyping Tactile Charts with Smart Defaults Mengzhu (Katie) Chen, Isabella Pedraza Pineros, Arvind Satyanarayan, Jonathan Zong Abstract . Tactile charts are essential for conveying data to blind and low vision (BLV) readers but are difficult for designers to construct. Non-expert designers face barriers to entry due to complex guidelines, while experts struggle with fragmented and time-consuming workflows that involve extensive customization. Inspired by formative interviews with expert tactile graphics designers, we created Tactile Vega-Lite (TVL): an extension of Vega-Lite that offers tactile-specific abstractions and synthesizes existing guidelines into a series of smart defaults. Predefined stylistic choices enable non-experts to produce guideline-compliant tactile charts quickly. Expert users can override defaults to tailor customizations for their intended audience. In a user study with 12 tactile graphics creators, we show that Tactile Vega-Lite enhances flexibility and consistency by automating tasks like adjusting spacing and translating braille while accelerating iterations through pre-defined textures and line styles. Through expert critique, we also learn more about tactile chart design best practices and design decisions. Conference: CHI Conference on Human Factors in Computing Systems; April 26-May 1, 2025; Yokohama, Japan CHI Conference on Human Factors in Computing Systems (CHI’25), April 26-May 1, 2025, Yokohama, Japan DOI: 10.1145/3706598.3714132 ISBN: 979-8-4007-1394-1/25/04 CCS: Human-centered computing Visualization systems and tools CCS: Human-centered computing Accessibility systems and tools † † cc-license: by 1. Introduction Tactile charts are an essential tool for blind and low vision (BLV) people to independently explore data and participate equally in discussions involving statistical analysis ( Engel and Weber, 2019 ; Watanabe et al., 2012 ) . As a subset of tactile graphics, tactile charts use braille, raised lines, textured areas, and distinct marks to represent statistical data through touch ( Engel and Weber, 2017b ) . Research has found that tactile charts facilitate independent data exploration, communicate spatial relationships, provide non-sequential access to information, and support content comprehension and memorization ( Engel and Weber, 2019 ) . They surpass printed or electronic tables in helping blind users understand the correlations between variables ( Watanabe et al., 2012 ) , helping students learn statistical concepts, and developing tactile graphic literacy ( Engel and Weber, 2017b ; Aldrich et al., 2003 ) . However, when we conducted formative interviews with tactile graphics designers ( section 3 ), we found that the existing workflows for designing and prototyping tactile charts are, unfortunately, tedious and time-consuming, requiring a high level of skill and experience. We found that because best practices for tactile design differ significantly from practices in visual design, non-expert designers face significant barriers to entry. For example, in visual design, it is common to optimize for space efficiency or aesthetics by orienting text sideways or scaling it while still maintaining legibility. However, this practice does not translate well to tactile design. Braille, the standard form of text in tactile charts, cannot be scaled, and orienting it in different directions can make it impossible to read. Prioritizing space efficiency can lead to overcrowding, which hinders the clarity and usability of tactile charts, especially for non-expert designers. Existing guidelines for tactile graphic design have a steep learning curve and can be overwhelming for those new to the field. Even for expert tactile designers, difficulties arise due to a lack of dedicated software tools for creating tactile charts. Designers typically need to master multiple tools to create a single chart, including general-purpose vector drawing tools and braille translation software. Each has its own learning curve, and designers must repeat tedious operations, such as transferring designs between tools, to make manual adjustments, even for minor modifications. The labor-intensive, specialized work involved in this process hinders the adoption of tactile charts in practice. To address these challenges, we present Tactile Vega-Lite (TVL): extensions to Vega-Lite ( Satyanarayan et al., 2017 ) for rapidly prototyping tactile charts. In contrast to existing workflows that rely on fragmented functionality across various general-purpose tools, TVL introduces a set of domain-specific abstractions for tactile chart design ( section 4 ). These include support for tactile encoding channels, braille, navigational aids, and layout configuration. These extensions to Vega-Lite enable designers to benefit from the advantages of existing visualization grammars—such as the ability to systematically enumerate a design space—while still expressing tactile-specific affordances. They also enable authors of Vega-Lite visualizations to make small edits to existing visual specifications to create tactile charts without requiring deep tactile design expertise. Just as Vega-Lite provides smart defaults that enable authors to easily specify common visual chart forms, TVL’s defaults generate tactile charts that align with established guidelines. For example, TVL renders grid lines as the least tactually distinct lines on a graph, as recommended by the Braille Authority of North America (BANA) guidelines ( Braille Authority of North America, 2022 ) . Also, TVL automatically places the x-axis title below the labels and aligns it to the leftmost x-axis label, where readers conventionally expect to find it. Expert designers can override these defaults to make context-specific adjustments for their output medium (e.g. braille embosser, swell paper) or intended audience. Similarly, predefined palettes of lines and area textures allow non-experts to choose from a set of reliable options while enabling experts to rapidly prototype with reusable texture configurations. We implement a prototype TVL editor that takes a declarative TVL specification and outputs a tactile chart in SVG format. These SVGs can be printed using various production methods; popular options include an embosser or a swell form machine. Through an example gallery ( section 5 ), we demonstrate that TVL expresses a variety of chart forms and layouts. We also conducted a user study ( section 6 ) with 12 tactile graphics designers to understand how TVL compares to their existing design workflows. We found that participants valued predefined textures and line styles for consistency and efficiency, aligning with existing workflows. Customizability was crucial for adapting tactile charts, especially for younger readers who benefit from grid lines and clear spacing. Differences in expectations between professional designers and Teachers of Students with Visual Impairments (TVIs) 1 1 1 The term “Teacher of Students with Visual Impairments” (TVI) is widely used by organizations such as the American Printing House for the Blind (APH) and the Braille Authority of North America (BANA) to refer to educators who work with blind and low vision students. We acknowledge that the term “visually impaired” can be considered harmful language ( Ather Sharif, 2024 ) , or otherwise not preferred ( Hanson et al., 2015 ) . In this paper, we avoid using this term to describe people. However, we use TVI for clarity when referencing the specific job title, to maintain consistency with its use by educational institutions like APH ( Garcia-Mejia, 2024 ) , by educators ( Carmen Willings, 2020 ) , and by the state in licensing teachers ( Anthony and Adams, 2005 ) . revealed a trade-off between advanced design capabilities and practical, easy-to-use solutions, with professionals seeking more complex tools and TVIs favoring simplicity and efficiency. Through an expert critique of the system’s defaults, we found that designers appreciated the tactile-specific design assets, such as predefined textures and line styles, which encouraged reasoning and prototyping to determine the optimal design for different audiences. This feedback